A design method for fuel molecules

Abstract

The present invention discloses a method for designing fuel molecules. The disclosed method involves screening using density, volumetric calorific value, melting point, flash point at normal pressure, specific impulse, and synthesis difficulty to select fuel molecules with excellent performance and a minimum number of synthesis steps. Based on molecules that meet all screening criteria, all non-ring atoms on the material are trimmed and the fragments are combined with molecules that do not contain non-ring atoms to design new fuel molecules. Finally, based on the screening criteria, target fuel molecules with excellent performance and a minimum number of synthesis steps are selected from these molecules. This method overcomes the shortcomings of traditional methods, such as small design sample sizes, molecular structures limited by the designer's knowledge, and a lack of consideration for the feasibility of molecular synthesis. It can shorten the time required for fuel molecule design screening and improve fuel molecule design efficiency.

Description

Technical Field

[0001] The present invention relates to a fuel molecule design method, in particular to a fuel molecule design method. Background Art

[0002] Research on aircraft fuel design is a fundamental area of focus in aerospace technology. It directly impacts aircraft power, thus directly influencing key performance indicators such as range, speed, and payload. To improve aircraft power, given the limited fuel tank capacity, fuels with high density and volumetric calorific value are generally required. Therefore, it is crucial to conduct research on fuel molecule design to accelerate the development of high-performance fuel molecules.

[0003] Existing technologies mainly rely on the chemical intuition of experimental scientists. According to the characteristics of existing fuel molecules, several fuel molecular structures are constructed through subjective understanding to achieve the design of fuel molecules. Although this technology may obtain target molecules with good performance, it is very dependent on the designers' long-term accumulated understanding of the relationship between fuel molecular structure and chemical properties. It requires a huge amount of time and manpower costs, which affects the speed of developing high-performance fuel molecules. In addition, due to the small sample size of designed molecules and the fact that most molecular structures are limited by the designers' inertial understanding, it is difficult for the obtained target molecules to break through the existing level in terms of performance. In addition, because the synthetic feasibility of the designed molecules was not considered during the design process, there was no assessment of whether the target molecules could be synthesized or the difficulty of synthesis. Therefore, in order to accelerate the development of high-performance fuel molecules, while taking into account the characteristics of the fuel molecules, it is necessary to use more effective design and screening strategies to be more likely to obtain new high-performance fuels with excellent comprehensive performance and experimental research value. Summary of the Invention

[0004] In view of the defects or shortcomings of the prior art, the present invention provides a fuel molecule design method.

[0005] To this end, the fuel molecule design method provided by the present invention includes:

[0006] Step 1: Construct an initial material database containing several material molecules, each with a density greater than 0.8 g / cm 3 , volume calorific value greater than 44.0MJ / L, melting point less than 273.15K, flash point greater than 56.0℃ at normal pressure, specific impulse performance parameter greater than 330s, and synthesis difficulty less than 4;

[0007] Step 2: For the materials containing non-cyclic atoms and the materials without non-cyclic atoms in the initial molecular database, the following processing is performed respectively: the non-cyclic atoms in the materials containing non-cyclic atoms are removed, and then the ring structure molecular fragments after the non-cyclic atoms are removed are stored in data set A; the material molecules without non-cyclic atoms in the initial molecular database are stored in data set B;

[0008] Step 3: construct multiple new molecules with different structures, where any new molecule is formed by connecting any hydrogen atom site on molecular fragment a with any hydrogen atom site on molecule b, where molecular fragment a is any molecular fragment in data set A, and material b is any material molecule in data set B;

[0009] Step 4: Calculate the density, volume calorific value, melting point, flash point at normal pressure, specific impulse and synthesis difficulty of each new molecule; select the one with density greater than 0.8 g / cm 3 , a new molecule with a volume calorific value greater than 44.0MJ / L, a melting point less than 273.15K, a flash point greater than 56.0℃ at normal pressure, a specific impulse greater than 330s, and a synthesis difficulty less than 4 is used as the designed fuel molecule.

[0010] Preferably, the structures of the ring structure molecular fragments in the dataset A are different.

[0011] Preferably, the step 1 comprises:

[0012] S11, the selected material density must be greater than 0.8g / cm 3 materials;

[0013] S12, selecting a material with a volume calorific value greater than 44.0 MJ / L from the materials selected in S11;

[0014] S13, selecting a material with a melting point less than 273.15K from the materials selected in S12;

[0015] S14, select a material with a flash point greater than 56.0°C at normal pressure from the material selected in S13;

[0016] S15, selecting a material having a specific impulse performance parameter greater than 330s from the materials selected in S14;

[0017] S16, selecting materials with a synthesis difficulty less than 4 from the materials selected in S15 to form an initial material database.

[0018] Further, step 1 in 10 6 Search and filter data at the 100-level and above to build an initial material database.

[0019] Optionally, the density, volume calorific value, melting point, flash point at normal pressure and specific impulse performance parameters are calculated using a group addition method.

[0020] The present invention uses density, volumetric calorific value, melting point, flash point at atmospheric pressure, specific impulse, and synthesis difficulty to screen and select fuel molecules with excellent performance and a minimum number of synthesis steps. Based on molecules that meet all screening criteria, all non-ring atoms in the material are trimmed and the fragments are combined with molecules that do not contain non-ring atoms to design new fuel molecules. Finally, the target fuel molecule with excellent performance and a minimum number of synthesis steps is selected based on the screening criteria. This method overcomes the shortcomings of traditional methods, such as small design sample size, molecular structure limitations due to the designer's knowledge, and a lack of consideration of the feasibility of molecular synthesis. It can shorten the time required for fuel molecule design screening and improve fuel molecule design efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 Five target fuel molecular structures designed for embodiments of the present invention;

[0022] Figure 2 This is an example diagram of the steps of identifying and trimming non-in-ring atoms of the material according to an embodiment of the present invention;

[0023] Figure 3 2 is an example diagram of a combination design of a fragment and a molecule without non-ring atoms, shown in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] Unless otherwise specified, the scientific and technical terms used herein are understood according to the knowledge of ordinary technicians in the relevant fields.

[0025] The method of the present invention is based on molecular structure, so any molecular database (one that stores molecular structure data) is applicable to the present invention. The following examples use the GDB-13 database as an example, primarily because it stores saturated hydrocarbon molecules (which reduces the steps of element screening and chemical bond screening).

[0026] The group addition method is derived from the literature [J. Marrero; R. Gani, Group-contribution based estimation of pure component properties. Fluid Phase Equilib. 2001, 183-208].

[0027] The synthesis difficulty value is based on the SCScore method. The SCScore method is derived from the literature [Coley, CW; Rogers, L.; Green, WH; Jensen, KF, Scscore: Synthetic Complexity Learned from a Reaction Corpus. J. Chem. Inf. Model. 2018, 58, 252-261].

[0028] The following embodiments of the present invention are further described in detail in conjunction with the accompanying drawings and specific embodiments. It should be understood that the following embodiments or drawings are only used to illustrate the present invention and are not used to limit the scope of protection of the present invention.

[0029] Example:

[0030] This example uses steps 11-16 of the present invention to screen the GDB-13 database and obtains 5475 data;

[0031] The dataset was constructed using step 2 of the present invention. Finally, dataset B contained 1872 material molecules, and dataset A contained 5734 ring structure fragments.

[0032] Using the steps described in step 3, a total of 858723842 new molecules were constructed. 37896 fuel molecules that met the screening criteria were screened using the conditions described in step 4. Figure 1 , five examples of designed fuel molecules are given, and the properties of each fuel molecule are shown in Table 1 below.

[0033] Table 1

[0034]

[0035]

[0036] To improve work efficiency, the above scheme can be implemented using database-specific coding and open source software. If necessary, existing computer programming programs can be used to implement batch processing of data.

[0037] Specifically, the Smiles coding information of materials is read in batches from the GDB-13 database; then the molecular parameters are calculated using the coding and corresponding open source software;

[0038] The method for reading the material Smiles coding information is as follows: calling the subroutine pd.read_csv() in the open source software package Pandas to batch read the material Smiles coding information;

[0039] The density, melting point, volume calorific value, flash point at normal pressure, and specific impulse performance parameters are obtained by converting the Smiles code of a material into the Mol format structure of the material by calling the Chem.MolFromSmiles() subroutine in the open source software package RDKit; then programming to identify the number of each group of the material, and then calculating it by the group addition method;

[0040] The formation enthalpy data required for the calculation of the volume calorific value, flash point at normal pressure and specific impulse performance parameters are obtained by the open source software MOPAC program at PM7 calculation accuracy.

[0041] Take the molecule with Smiles code CC1CC2C3CCCC23C2CC12C in GDB-13 database as an example,

[0042] First, the smiles code is converted into a Mol format structure by the Chem.MolFromSmiles() subroutine; then the structure of each group is defined by Chem.MolFromSmarts(), and then the subroutine mol.GetsubstructMatches() is used to identify each group, and finally the number of each group (including CH3:2; CH2 cyc :5;CH cyc :4;C cyc :2;CH cyc -CH3:1;C cyc -CH3:1;CH multi-cyc :3;C multi-cyc :2;), wherein the subscript cyc indicates that the group belongs to a ring structure, and multi-cyc indicates that the group belongs to multiple ring structures;

[0043] The formation enthalpy of the material was obtained as 46.2 kJ / mol using the MOPAC program at PM7 accuracy. The density of the material was then obtained as 0.893 g / cm by the group addition method based on the formation enthalpy data and the number of groups. 3 , melting point 284.9K, volume calorific value 38.28MJ / L, flash point 77.9°C at atmospheric pressure, and specific impulse 340.9s. Specifically, the synthesis difficulty of a material is calculated in batches using the SCScore method, using a computer program to batch read the Smiles code of a material. This method model then calculates the final SCScore. For example, using a molecule with the Smiles code CC1CC2C3CCCC23C2CC12C in the GDB-13 database, the synthesis difficulty of this material is 3.508.

[0044] The specific solution using coding and open source software to implement step 2 is as follows:

[0045] S20: Read the Smiles code of a material selected in step 1;

[0046] S21: using the Chem.MolFromSmiles() subroutine in the open source software package RDKit to convert the Smiles code of the material into a Mol format structure;

[0047] S22: Identify non-ring atoms, specifically by using a computer program to traverse all atoms in the material;

[0048] S23: Determine whether the material has non-ring atoms;

[0049] S24-a: If the material does not have any non-ring atoms, store the Smiles code of the material into data set B;

[0050] S24-b-1: If the material contains non-cyclic atoms, the non-cyclic atoms in the material are trimmed. Specifically, the atom trimming can be achieved using a computer program;

[0051] S24-b-2: The Smiles encoding of the trimmed segments excluding non-ring atoms is stored in dataset A;

[0052] Example: See Figure 2 , read two molecules (molecule 1 and molecule 2) with Smiles codes of C1C2CC1C1C3C1C1CC3C3C2C13 and C1CC1C(C1CC1)(C1CC1)C1CC1 through Chem.MolFromSmiles(); then identify non-cyclic structures including ('[CH4]', '[CH3]', '[CH2;!R]', '[CH1;!R]', '[CH0;!R]') through Chem.MolFromSmarts(); finally, determine the number of the five non-cyclic structures. If the value is non-zero, it means that it contains non-cyclic atoms. Among them, molecule 1 does not have non-cyclic atoms and is stored in data set B; molecule 2 does have non-cyclic atoms, and then use MurckS caffold.GetScaffoldForMol() obtains the skeletal structure after trimming the branched groups (i.e., removing the '[CH3]' group; in this example, molecule 2 does not contain a CH3 group). SetIntProp() is then used to register the chemical bonds in the molecule. GetRingInfo() and BondRings() are used to obtain the serial numbers of the ring structure chemical bonds, and then the serial numbers of the non-ring structure bonds. FragmentOnBonds() is then used to trim the non-ring structure bonds to obtain the ring structure in molecule 2. Although molecule 2 contains four three-membered rings after trimming, since the three-membered ring structures are the same, only one three-membered ring is stored in dataset A, ensuring that the structures of the ring structure molecular fragments in dataset A are different.

[0053] In the specific scheme, the operations of implementing step 3 using coding and open source software are as follows:

[0054] S30-a: Using the open source software package RDKit and programming, read a segment from dataset A and generate a corresponding Mol format structure, denoted as Am;

[0055] S30-b: Using the open source software package RDKit and programming, read a material molecule from the dataset B and generate the corresponding Mol format structure, denoted as Bm;

[0056] S31-a: Using the open source software package RDKit, the hydrogen atoms on the fragment Am are explicitly represented, and then programming is performed to realize hydrogen atom recognition and labeling. The fragment Am is labeled as Am-*.

[0057] S31-b: Using the open source software package RDKit, the hydrogen atoms on the fragment Bm are explicitly represented, and then programming is performed to realize hydrogen atom recognition and labeling. The material Bm is labeled as Bm-*.

[0058] S32: defining a combination rule: Am-*+Bm-*→Cm, where Cm is a new molecule; designing a new molecule according to the combination rule, specifically using a computer program to design a combination of Am-* and Bm-* that satisfies the combination rule to generate a new material molecule;

[0059] Example: See Figure 3 , select a fragment from data set A and record it as Am, its Smiles code is C12C3C4C3C5CC1C6C2CC4C56, use Chem.AddHs() to display the hydrogen atoms, and then use ReplaceSubstructs to represent a hydrogen atom as * (identified as a connection site). Due to the symmetric structure of the fragment, there are only 7 different Am-*;

[0060] Select a molecule from dataset B, denoted as Bm, with Smiles encoding of C12CC1C34CC5C6C5C3C7C2C4C67. Use Chem.AddHs() to display hydrogen atoms, and then use ReplaceSubstructs to represent a hydrogen atom as * (identified as a connection site). Due to the asymmetric structure of this molecule, there are 13 different Bm-*. Then, according to the defined combination rule Am-*+Bm-*→Cm, Am-* and Bm-* in this example will generate a total of 91 new molecules.

[0061] The database-based high-performance fuel molecule design method of the present invention overcomes the shortcomings of traditional methods such as small design sample size, molecular structure limited by designers' knowledge, and consideration of molecular synthesis feasibility. It realizes the acquisition of fuel molecular structural property characteristics from the database and the design of new high-performance fuel molecules based on them.

[0062] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A fuel molecule design method, characterized in that: Methods include: Step 1: Construct an initial material database containing several material molecules, each with a density greater than 0.8 g / cm 3 , volume calorific value greater than 44.0MJ / L, melting point less than 273.15K, flash point greater than 56.0℃ at normal pressure, specific impulse performance parameter greater than 330s, and synthesis difficulty less than 4; Step 2: For the materials containing non-cyclic atoms and the materials without non-cyclic atoms in the initial material database, the following processing is performed respectively: the non-cyclic atoms in the materials containing non-cyclic atoms are removed, and then the ring structure molecular fragments after the non-cyclic atoms are removed are stored in data set A; the material molecules without non-cyclic atoms in the initial material database are stored in data set B; Step 3: construct multiple new molecules with different structures, where any new molecule is formed by connecting any hydrogen atom site on molecule fragment a with any hydrogen atom site on molecule b, where molecule fragment a is any molecular fragment in data set A, and molecule b is any material molecule in data set B; Step 4, respectively calculate the density, volume calorific value, melting point, flash point at normal pressure, specific impulse performance parameters and synthesis difficulty of each new molecule obtained in step 3; select the one with density greater than 0.8g / cm 3 , a new molecule with a volume calorific value greater than 44.0MJ / L, a melting point less than 273.15K, a flash point greater than 56.0℃ at normal pressure, a specific impulse performance parameter greater than 330s, and a synthesis difficulty less than 4 is designed as the fuel molecule.

2. The fuel molecule design method according to claim 1, wherein: The structures of the ring structure molecular fragments in the dataset A are different.

3. The fuel molecule design method according to claim 1, wherein: The step 1 comprises: S11, the selected material density must be greater than 0.8g / cm 3 materials; S12, selecting a material with a volume calorific value greater than 44.0 MJ / L from the materials selected in S11; S13, selecting a material with a melting point less than 273.15K from the materials selected in S12; S14, select a material with a flash point greater than 56.0°C at normal pressure from the material selected in S13; S15, selecting a material having a specific impulse performance parameter greater than 330s from the materials selected in S14; S16, selecting materials with a synthesis difficulty less than 4 from the materials selected in S15 to form an initial material database.

4. The fuel molecule design method according to claim 1 or 3, characterized in that: In the containing 10 6 Search and filter data at the 100-level and above to build an initial material database.

5. The fuel molecule design method according to claim 1, wherein: The density, volume calorific value, melting point, flash point at normal pressure and specific impulse performance parameters are calculated using the group addition method.

Images