A high-throughput computational screening method for bifunctional MOFs for catalytic separation

Abstract

The present invention discloses a high-throughput computational screening method for catalytic separation dual-function MOFs, belonging to the field of computational chemistry and nanocomposite catalytic materials. The screening method is to accurately and efficiently screen out promising MOFs materials with catalytic separation dual functions from the MOFs candidate pool through high-throughput calculations of multi-level theories step by step. The advantages of the present invention are: 1) the use of high-throughput parallel calculations effectively shortens the data acquisition time; 2) the use of high-throughput step-by-step calculations avoids the waste of resources caused by traditional experimentally intensive "trial and error" experiments, reduces research and development costs, and shortens the research and development cycle; 3) the screening method provided by the present invention takes into account both catalytic activity and the timely separation of by-products, and has good practical value.

Description

Technical Field

[0001] The invention relates to the technical field of computational chemistry and nanocomposite catalytic materials, and in particular to a high-throughput computational screening method for catalytic separation of bifunctional MOFs. Background Art

[0002] The efficient conversion and utilization of CO2 is of great strategic significance for alleviating the energy crisis and achieving the goal of "carbon neutrality". CO2 can be converted into a variety of high-value-added chemicals and fuels, such as carbon monoxide, methanol, methane, formic acid, etc. Among them, the product methanol is called liquid sunshine and can serve as a green liquid fuel. Compared with energy carriers such as hydrogen energy and lithium-ion batteries, it has outstanding advantages in terms of volume energy density, mass energy density, transportation cost, and infrastructure investment. In addition, methanol is also an important organic chemical raw material, and its consumption is second only to ethylene, propylene and benzene. Therefore, CO2 hydrogenation to methanol is one of the most promising high-value conversion technologies for CO2.

[0003] In the CO2 hydrogenation to methanol reaction system, the main reactions are:

[0004]

[0005]

[0006]

[0007] Reactions (1) and (2) are competitive reactions. From a thermodynamic point of view, low temperature is favorable for the hydrogenation of CO2 to produce methanol. However, the C=O bond energy of CO2 is high (750 kJ / mol), and it is difficult to activate at low temperature. At the same time, reactions (1) and (3) are reactions in which the number of moles decreases. High pressure is favorable for the preparation of methanol, but the cost is high and there are certain safety risks.

[0008] Catalytic materials can effectively regulate the CO2 hydrogenation path and improve methanol selectivity, so the core of CO2 hydrogenation to methanol is to develop efficient catalytic materials. In addition, the products of hydrogenation (methanol and water) cannot be discharged in time, which hinders the continuous progress of the reaction. Therefore, this area urgently needs to develop a bifunctional catalytic material with excellent catalytic performance of CO2 hydrogenation and can directional separation of hydrogenation products, in situ separation of hydrogenation products during CO2 hydrogenation, so that CO2 hydrogenation to methanol reaction efficiency approaches thermodynamic limit and chemical reaction balance is constantly shifted to the right under specific temperature and pressure conditions. In addition, the in-situ removal of byproduct water can also take away a part of the heat generated by the exothermic reaction, avoiding the aggregation, sintering and even deactivation problems of the catalyst caused by byproduct steam and excessive temperature.

[0009] Metal-organic frameworks (MOFs), also known as porous coordination polymers or porous coordination networks, are three-dimensional porous materials with regular structures that are assembled according to a certain topological structure with metals (atoms, clusters, etc.) as nodes and organic molecules as ligands. Due to their comprehensive favorable properties such as large surface area, high porosity, adjustable pore size, high atomic utilization and shape selectivity, MOFs are considered to be ideal candidate materials that can catalyze the hydrogenation of carbon dioxide to methanol and effectively separate water. However, the high flexibility of MOFs composition and structure makes their number extremely large, and the traditional experimental "trial and error" research method is almost ineffective. Therefore, the development of efficient screening methods is of great significance for the research and development of high-performance bifunctional MOFs.

[0010] Theoretical calculations are mainly based on density functional theory (DFT), molecular dynamics (MD) and Monte Carlo method (MC), which can simulate material properties with considerable accuracy. With the continuous improvement of computer computing power, theoretical simulation of materials has become feasible, especially in the field of catalysis, playing an irreplaceable role in analyzing catalytic mechanisms and structural evolution.

[0011] However, for large-scale materials with variable structures such as MOFs, current theoretical simulations can only reveal the catalytic reaction mechanism of a few MOFs, which is a means of understanding the material after the material is synthesized, and its huge potential in guiding experimental synthesis has not yet been fully explored. "Calculation first, optimal synthesis" is the only way to develop high-performance bifunctional MOFs, which can effectively avoid the waste of energy and resources caused by blind trial and error, thereby greatly improving the research and development efficiency of MOFs materials. Therefore, the development of a high-throughput screening and catalytic separation bifunctional MOFs material method has broad practical prospects. Summary of the invention

[0012] The technical problem to be solved by the present invention is to provide a high-throughput computational screening method for catalytic separation bifunctional MOFs, which can accurately and efficiently screen out promising MOFs materials with catalytic separation bifunctionality from the MOFs candidate pool through high-throughput calculations of multi-level theories (density functional theory (DFT), molecular dynamics (MD), Monte Carlo (MC)), effectively guide the experimental synthesis of MOFs materials, and achieve a significant leap in research and development efficiency.

[0013] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0014] A high-throughput computational screening method for catalytic separation bifunctional MOFs comprises the following steps:

[0015] S1. Extract the secondary structural units (SBUs) and organic ligands of MOFs from the CoRE (Computation-Ready, Experimental) database as building blocks, extract the topological structure from the RCSR (Reticular Chemistry Structure Resource) database as the structural skeleton of MOFs, and construct a MOFs structural model with high throughput as the structural data set to be screened;

[0016] S2. Molecular dynamics (MD) is used to optimize the structure of the MOFs structure model obtained in S1 at the molecular level, and the geometric structure parameters of the MOFs material are obtained by Voronoi decomposition. Then, MOFs materials whose pore restriction diameter in the geometric structure parameters is larger than the van der Waals diameter of methanol are screened for subsequent screening.

[0017] S3, using MD and MC to analyze the physical adsorption and diffusion behaviors of CO2 and H2O in the pores of the MOFs materials screened in S2, and screen MOFs materials that can selectively allow H2O and methanol to pass through for subsequent screening;

[0018] S4. Using the DFT (Density Functional Theory) method combined with the Complete LST / QST method, we searched for MOFs materials with good CO2 chemical adsorption and desorption performance and possible transition states in the catalytic CO2 hydrogenation process among the MOFs materials screened in S3. We also used the finite displacement method to calculate the Gibbs free energy (G) of the relevant configurations of the MOFs materials, analyzed the reaction paths of catalytic CO2 hydrogenation of different MOFs materials, and the activation energy barriers of CO2 hydrogenation to methanol, and finally obtained catalytic separation bifunctional MOFs materials.

[0019] Among them, preferably, in S1, the conditions for high-throughput assembly include: a preset bond length of 1.54 to 2.0 angstroms, a bond perturbation coefficient of 4.0 to 6.0, a pre-scaling coefficient of 0.8 to 1.2, single metal atoms are allowed as nodes, the optimization method is L-BFGS-B, and pseudoatoms are allowed to be removed.

[0020] Preferably, in S2, the conditions for structural optimization at the molecular level include: the force field adopts the universal force field, the charge is calculated by the Qeq method, the integral of the electrostatic interaction adopts the PPPM method, the integral of the van der Waals interaction adopts the Ewald method, and the convergence thresholds of energy, force and displacement in the structural optimization are respectively and Structural optimization optimizes lattice constants, and the algorithm uses one of the steepest descent method, conjugate gradient method, and quasi-Newton method according to the convergence situation.

[0021] Wherein, preferably, in S3, the analysis conditions of the physical adsorption and diffusion behavior include:

[0022] Physical adsorption was simulated using the Monte Carlo (MC) method. The specific conditions include: the method is the Grand Canonical Monte Carlo method, the force field is the general force field, the charge is calculated using the Qeq method, the electrostatic integration is the Ewald method, the van der Waals interaction integration is the Group based method, the sampling interval is 25 steps, and the integration grid interval is

[0023] The diffusion was simulated by molecular dynamics method. The specific conditions included: using NVT ensemble, setting the temperature to the temperature required for the reaction, the dynamics simulation time was 500ps, using Nose insulator, using general force field, using Qeq method to calculate the charge, using PPPM method for the integral of electrostatic interaction, and using Ewald method for the integral of van der Waals interaction.

[0024] Wherein, preferably, in S4, the conditions for analyzing the reaction path of the catalytic CO2 hydrogenation and the activation energy barrier of CO2 hydrogenation to methanol include: using the DFT method, using the GGA-RPBE exchange-correlation functional, and all calculations are performed at the same accuracy; the default value set by the valence electron selection functional, the pseudopotential uses an ultrasoft pseudopotential, the cutoff energy is selected between 550eV and 800eV according to the type of elements contained in the system, the Monkhorst-Pack lattice sampling accuracy is set to high, the self-consistent convergence threshold is 1.0e-7 / atom to 1.0e-5 / atom, and the convergence thresholds of energy, force, maximum lattice stress and displacement in the structural optimization are 5.0e-6eV / atom to 1.0e-5eV / atom, respectively. 0.02~0.2GPa and

[0025] Among them, preferably, in S2, the geometric structure parameters include maximum pore diameter (LCD), pore limiting diameter (PLD), specific surface area (SSA), porosity (VF) and the like.

[0026] Among them, in S1, the MOFs structure in the MOFs structural model includes a MOFs material synthesized from several soluble metal salts and organic ligands.

[0027] Exemplarily, the several soluble metal salts include: one or more selected from nitrates, chlorides, phosphates, sulfates, acetates, etc. of niobium, titanium, vanadium, molybdenum, palladium, chromium, silver, manganese, nickel, iron, tin, cobalt, tungsten, copper, zinc, zirconium, and aluminum.

[0028] Exemplarily, the organic ligand includes: one or more of 1,4-benzenedicarboxylic acid, 1,3,5-benzenetricarboxylic acid, 1,2-benzenedicarboxylic acid, 1,2,4,5-benzenetetracarboxylic acid, benzenehexacarboxylic acid, 2-sulfonateterephthalic acid, 2-nitroterephthalic acid, 2-aminoterephthalic acid, 1,1':4',1"-phenyl-4,4"-dicarboxylic acid, 1,1'-diphenyl-4,4'-dicarboxylic acid, piperazine, pyrazine, dimethylimidazole, triethylenediamine, 4,4'-bipyridine, 1,3-di(4-pyridine)propane, etc.

[0029] The beneficial effects of the above technical solution of the present invention are as follows:

[0030] 1) High-throughput parallel computing is used to effectively shorten the data acquisition time;

[0031] 2) The high-throughput step-by-step calculation method is adopted to avoid the waste of resources caused by traditional experimental-intensive "trial and error" experiments, reduce R&D costs, and shorten the R&D cycle;

[0032] 3) The screening method provided by the present invention takes into account both the catalytic activity and the timely separation of by-products, and has good practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 Schematic diagram of several MOFs structures obtained by high-throughput assembly in the method of the present invention;

[0034] Figure 2 The basic setting parameter page, electronic item setting page, and property parameter setting page for high-throughput calculation settings in the method described in the present invention. DETAILED DESCRIPTION

[0035] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0036] Example 1

[0037] (1) High-throughput modeling and molecular-level structural optimization of two-dimensional MOFs:

[0038] The 2D topologies of sql, hcb, kgm, etc. in the RCSR database are used as the MOFs skeleton, and the secondary structural units containing metals such as Zr and Cu in the CoRE database are extracted and used together with terephthalic acid as building blocks to participate in the construction of the MOFs structure. The preset bond length is 1.54 angstroms, the perturbation coefficient of the bond is 5.0, the pre-scaling coefficient is 1.0, single metal atoms are allowed as nodes, the optimization method is L-BFGS-B, and pseudoatoms are allowed to be removed. Several MOFs structures constructed are as follows Figure 1 shown.

[0039] The optimization at the molecular level uses the universal force field, the Qeq method is used to calculate the charge, the PPPM method is used for the integration of electrostatic interactions, and the Ewald method is used for the integration of van der Waals interactions. The convergence thresholds of energy, force, and displacement in structural optimization are 2.0e-5 kcal / mol, and Structural optimization optimizes lattice constants and uses the conjugate gradient method.

[0040] (2) High-throughput step-by-step screening of MOFs materials:

[0041] The optimized MOFs structure configuration was analyzed by the Voronoi decomposition method to obtain the geometric structural parameters of the MOFs material, such as the maximum pore size (LCD), pore limiting diameter (PLD), specific surface area (SSA), and porosity (VF). Then, MOFs materials with pore limiting diameters greater than the van der Waals diameter of methanol in the geometric structural parameters were screened for subsequent screening.

[0042] MD and MC were used to analyze the physical adsorption and diffusion behavior of CO2 and H2O in the pores of the above-screened MOFs materials, and MOFs materials that can selectively allow H2O and methanol to pass through were screened for subsequent screening. Among them, physical adsorption was simulated by MC method, the method was the Grand Canonical Monte Carlo method, the force field was the universal force field, the charge was calculated by Qeq method, the integral of electrostatic interaction was the Ewald method, the integral of van der Waals interaction was the Group based method, the sampling interval was 25 steps, and the integral grid interval was The diffusion was calculated using the MD method, the NVT ensemble, the temperature was set to the temperature required for the reaction, the kinetic simulation time was 500 ps, ​​the Nose insulator was used, the universal force field was used as the force field, the Qeq method was used to calculate the charge, the PPPM method was used for the integration of electrostatic interactions, and the Ewald method was used for the integration of van der Waals interactions.

[0043] Using the DFT method (parameter settings such as Figure 2), using the GGA-RPBE exchange-correlation functional, all calculations were performed at the same accuracy. The default values ​​of the valence electron selection functional were used, the pseudopotential was an ultrasoft pseudopotential, the cutoff energy was 650 eV, the Monkhorst-Pack grid sampling accuracy was set to high (4×4×1), the self-consistent convergence threshold was 1.0e-6 / atom, and the convergence thresholds of energy, force, maximum lattice stress, and displacement in the structural optimization were 1.0e-5 eV / atom, 0.05GPa and To analyze the reaction pathway of catalytic CO2 hydrogenation of different MOFs materials and the activation energy barrier of CO2 hydrogenation to methanol.

[0044] The complete linear stepwise and quadratic stepwise (Complete LST / QST) methods were used to search for MOFs materials with good CO2 chemical adsorption and desorption performance and possible transition states in the CO2 hydrogenation process. The finite displacement method was used to calculate the Gibbs free energy of the relevant configurations, and finally the catalytic separation bifunctional MOFs materials were obtained, including Figure 1 a etc. Then the experimental synthesis was carried out, and the test results showed that the material obtained by this method was in good agreement with the experimentally synthesized material.

[0045] Example 2

[0046] (1) High-throughput modeling and molecular-level structural optimization of three-dimensional MOFs:

[0047] The tbo topology in the RCSR database was used as the MOFs skeleton, and the secondary structural units containing metals such as Zr and Cu in the CoRE database were extracted and used together with trimesic acid as building blocks to participate in the construction of the MOFs structure. The preset bond length was 1.54 angstroms, the bond perturbation coefficient was 5.0, the pre-scaling coefficient was 1.0, single metal atoms were allowed to be used as nodes, the optimization method was L-BFGS-B, and pseudoatoms were allowed to be removed. The MOFs constructed were as follows Figure 1 shown.

[0048] The optimization at the molecular level uses the universal force field, the Qeq method is used to calculate the charge, the PPPM method is used for the integration of electrostatic interactions, and the Ewald method is used for the integration of van der Waals interactions. The convergence thresholds of energy, force, and displacement in structural optimization are 2.0e-5 kcal / mol, and Structural optimization optimizes lattice constants and uses the conjugate gradient method.

[0049] (2) High-throughput step-by-step screening of MOFs materials:

[0050] The optimized MOFs structure configuration was analyzed by the Voronoi decomposition method to obtain the geometric structural parameters of the MOFs material, such as the maximum pore size (LCD), pore limiting diameter (PLD), specific surface area (SSA), and porosity (VF). Then, MOFs materials with pore limiting diameters greater than the van der Waals diameter of methanol in the geometric structural parameters were screened for subsequent screening.

[0051] MD and MC were used to analyze the physical adsorption and diffusion behavior of CO2 and H2O in the pores of the above-screened MOFs materials, and MOFs materials that can selectively allow H2O and methanol to pass through were screened for subsequent screening. Among them, physical adsorption was simulated by MC method, the method was the Grand Canonical Monte Carlo method, the force field was the universal force field, the charge was calculated by Qeq method, the integral of electrostatic interaction was the Ewald method, the integral of van der Waals interaction was the Group based method, the sampling interval was 25 steps, and the integral grid interval was The diffusion was calculated using the MD method, the NVT ensemble, the temperature was set to the temperature required for the reaction, the kinetic simulation time was 500 ps, ​​the Nose insulator was used, the universal force field was used as the force field, the Qeq method was used to calculate the charge, the PPPM method was used for the integration of electrostatic interactions, and the Ewald method was used for the integration of van der Waals interactions.

[0052] Using the DFT method (parameter settings such as Figure 2 ), using the GGA-RPBE exchange-correlation functional, all calculations were performed at the same accuracy. The default values ​​of the valence electron selection functional were used, the ultrasoft pseudopotential was used, the cutoff energy was selected as 700eV, the Monkhorst-Pack grid sampling accuracy was set to high (6×6×6), the self-consistent convergence threshold was 1.0e-7 / atom, and the convergence thresholds of energy, force, maximum lattice stress, and displacement in the structural optimization were 1.0e-5eV / atom, 0.05GPa and To analyze the reaction pathway of catalytic CO2 hydrogenation of different MOFs materials and the activation energy barrier of CO2 hydrogenation to methanol.

[0053] The Complete LST / QST method was used to search for MOFs materials with good CO2 chemical adsorption and desorption performance and possible transition states in the CO2 hydrogenation process, and the finite displacement method was used to calculate the Gibbs free energy of the relevant configurations. Finally, a catalytic separation bifunctional MOFs material was obtained. After that, an experimental synthesis was carried out, and the test results showed that the material obtained by this method was in good agreement with the experimentally synthesized material.

[0054] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A high-throughput computational screening method for catalytic separation bifunctional MOFs, characterized in that: The following steps are involved: S1. Extract the secondary structural units and organic ligands of MOFs from the CoRE database as building blocks, extract the topological structure from the RCSR database as the structural skeleton of MOFs, and construct a MOFs structural model in high throughput as the structural data set to be screened; In S1, the conditions for high-throughput assembly include: preset bond length of 1.54 to 2.0 angstroms, bond perturbation coefficient of 4.0 to 6.0, pre-scaling coefficient of 0.8 to 1.2, allowing single metal atoms as nodes, optimization method of L-BFGS-B, and allowing removal of pseudoatoms; S2. Use molecular dynamics to optimize the structure of the MOFs structure model obtained in S1 at the molecular level, and use the Voronoi decomposition method to obtain the geometric structure parameters of the MOFs material, and then screen the MOFs materials whose pore restriction diameter in the geometric structure parameters is larger than the van der Waals diameter of methanol for subsequent screening; In S2, the conditions for structural optimization at the molecular level include: the force field uses the general force field, the charge is calculated using the Qeq method, the integral of the electrostatic interaction uses the PPPM method, the integral of the van der Waals interaction uses the Ewald method, and the convergence thresholds of energy, force, and displacement in the structural optimization are and Structural optimization optimizes the lattice constants, and the algorithm uses one of the steepest descent method, conjugate gradient method, and quasi-Newton method according to the convergence situation; In S2, the geometric structure parameters include maximum pore size, pore limiting diameter, specific surface area, and porosity; S3, using molecular dynamics and Monte Carlo to analyze the physical adsorption and diffusion behaviors of CO2 and H2O in the pores of the MOFs materials screened in S2, to screen MOFs materials that can selectively allow H2O and methanol to pass through, and to conduct subsequent screening; In S3, the analysis conditions of the physical adsorption and diffusion behavior include: Physical adsorption was simulated using the Monte Carlo method. The specific conditions include: the method is the Grand Canonical Monte Carlo method, the force field is the general force field, the charge is calculated using the Qeq method, the integral of the electrostatic interaction is the Ewald method, the integral of the van der Waals interaction is the Group based method, the sampling interval is 25 steps, and the integral grid interval is Diffusion was simulated by molecular dynamics method. The specific conditions include: NVT ensemble, temperature was set to the temperature required for reaction, dynamic simulation time was 500ps, Nose incubator was used, general force field was used for force field, Qeq method was used to calculate charge, PPPM method was used for integration of electrostatic interaction, and Ewald method was used for integration of van der Waals interaction. S4. Using DFT combined with the fully linear synchronous method and the quadratic synchronous method, we searched for MOFs materials with good CO2 chemical adsorption and desorption performance and possible transition states in the catalytic CO2 hydrogenation process among the MOFs materials screened in S3, and used the finite displacement method to calculate the Gibbs free energy of the relevant configurations of MOFs materials, and analyzed the reaction paths of catalytic CO2 hydrogenation of different MOFs materials and the activation energy barriers of CO2 hydrogenation to methanol, and finally obtained catalytic separation bifunctional MOFs materials; In S4, the conditions for analyzing the reaction path of catalytic CO2 hydrogenation and the activation energy barrier of CO2 hydrogenation to methanol include: using the DFT method, using the GGA-RPBE exchange-correlation functional, and all calculations are performed at the same accuracy; the default value set by the valence electron selection functional, the pseudopotential uses an ultrasoft pseudopotential, the cutoff energy is selected between 550eV and 800eV according to the type of elements contained in the system, the Monkhorst-Pack lattice sampling accuracy is set to high, the self-consistent convergence threshold is 1.0e-7 / atom to 1.0e-5 / atom, and the convergence thresholds of energy, force, maximum lattice stress, and displacement in structural optimization are 5.0e-6eV / atom to 1.0e-5eV / atom, respectively. 0.02~0.2GPa and 2. The high-throughput computational screening method according to claim 1, characterized in that: In S1, the MOFs structure in the MOFs structure model includes a MOFs material synthesized from a soluble metal salt and an organic ligand.

3. The high-throughput computational screening method according to claim 2, characterized in that: The several soluble metal salts include: one or more selected from nitrates, chlorides, phosphates, sulfates and acetates of niobium, titanium, vanadium, molybdenum, palladium, chromium, silver, manganese, nickel, iron, tin, cobalt, tungsten, copper, zinc, zirconium and aluminum.

4. The high-throughput computational screening method according to claim 2, characterized in that: The organic ligand includes: one or more of 1,4-phthalic acid, 1,3,5-benzenetricarboxylic acid, 1,2-phthalic acid, 1,2,4,5-benzenetetracarboxylic acid, mellitic acid, 2-sulfonateterephthalic acid, 2-nitroterephthalic acid, 2-aminoterephthalic acid, 1,1':4',1"-phenyl-4,4"-dicarboxylic acid, 1,1'-diphenyl-4,4'-dicarboxylic acid, piperazine, pyrazine, dimethylimidazole, triethylenediamine, 4,4'-bipyridine, and 1,3-di(4-pyridine)propane.

Images