A kind of kinetic calculation method for estimating intrinsic activity of alkane dehydrogenation to olefin in molecular sieve

Abstract

The present application belongs to the field of theoretical catalytic calculation, and particularly relates to a kind of kinetic calculation method for estimating the dehydrogenation activity of zeolite catalyst to alkane.The method comprises the following steps: 1) establishing a network of catalytic elementary reactions, considering the existence of intermediate radicals;2) DFT calculation of the adsorption energy of main species, thereby obtaining the reaction energy of each elementary reaction; at the same time, the activation energy barrier of each elementary reaction is calculated by DFT;3) constructing a function covering the adsorption energy and / or activation energy barrier;4) in the radical pressure prediction model, first guess the radical pressure, and according to the value of the radical pressure, carry out micro-kinetic simulation including the self-consistent model of coverage degree.The present application combines the characteristics of reaction mechanism, coverage effect and radical pressure in the pore for micro-kinetic simulation; can accurately predict the intrinsic activity of alkane dehydrogenation in zeolite, and realize the agreement of theoretical results and experimental results.

Description

Technical Field

[0001] This invention belongs to the field of theoretical catalysis calculation, specifically involving a kinetic calculation method for estimating the activity of molecular sieve catalysts for alkane dehydrogenation. Background Technology

[0002] Calculating the energy data (including the reaction free energy and activation energy) of each elementary reaction in alkane dehydrogenation within a molecular sieve catalyst using density functional theory (DFT) and then combining this with microkinetic simulations to calculate the reaction rate is a crucial method in theoretical catalysis for estimating catalytic activity. Therefore, the accuracy of catalytic activity calculations using microkinetic simulations is key to the widespread application of this method. However, in some specialized catalytic systems (such as molecular sieves and other catalysts with nanoscale pores), the reaction rates obtained from traditional DFT-based energy-based microkinetic simulations exhibit significant errors (greater than five orders of magnitude) compared to actual experimental results.

[0003] To further reduce the error in activity prediction, researchers have studied and improved the models for DFT calculations and microdynamic simulations from multiple perspectives. For example, Guo, Norskov, and other researchers conducted a series of studies on the influence of surface coverage in heterogeneous catalysts (J. Catal. 2019, 379, 52-59), and they believe that considering the effect of coverage (i.e., the coexistence state between adsorbates affects the energy in the current environment) in microdynamic simulations is a key factor in improving the accuracy of microdynamic simulations. Hu et al. (J. Comput. Chem. 2021, 42, 379-91) developed the powerful multi-scale microdynamic modeling software CATKINAS to reduce the error in the mathematical solution of multidimensional nonlinear equations in microdynamic simulations. However, applying these methods to the rate simulation of alkane dehydrogenation in molecular sieves still results in significant errors. The main causes of these errors are:

[0004] 1. The reaction mechanism for alkane dehydrogenation in the molecular sieve system is not accurately described, i.e., the elementary reaction involving free radicals is not considered (J. Am. Chem. Soc. 2020, 142, 16429-16436).

[0005] 2. Unlike the continuous influence of the coverage effect in traditional surface models, the coverage effect in molecular sieve systems is significantly different from that of traditional surfaces due to the independence of reactive sites.

[0006] 3. When considering the participation of free radicals in a reaction, the state of the free radicals is neither the "adsorbed state" on the active site nor the "gas phase" where the reactants / products are located. Therefore, the concentration of free radicals, an important parameter in microscopic kinetic simulations, is difficult to describe accurately. Summary of the Invention

[0007] This invention aims to solve the aforementioned problems by providing a kinetic calculation method for estimating the intrinsic activity of alkane dehydrogenation to olefins in molecular sieves. It utilizes microscopic kinetic simulation by combining reaction mechanisms, coverage effects, and free radical pressures within the pores; this method can accurately predict the intrinsic activity of alkane dehydrogenation in molecular sieves, achieving a good match between theoretical and experimental results.

[0008] The technical solution of the present invention to solve the above problems is as follows:

[0009] A kinetic calculation method for estimating the intrinsic activity of alkane dehydrogenation to olefins in molecular sieves includes the following steps:

[0010] 1) Establish a network for catalytic elementary reactions, taking into account the presence of intermediate free radicals;

[0011] 2) DFT calculation of the adsorption energy of the main species, thereby obtaining the reaction energy of each elementary reaction; at the same time, DFT calculation of the activation energy barrier of each elementary reaction.

[0012] 3) DFT calculation of adsorption energy and activation energy barrier of each species in molecular sieve at different coverage of major adsorbates at discrete sites; continuous processing of energy at different coverage at discontinuous sites to construct a function of coverage with adsorption energy and / or activation energy barrier;

[0013] 4) In the free radical pressure prediction model, the free radical pressure is first initially estimated, and then a micro-dynamic simulation including a coverage self-consistent model is performed based on the value of the free radical pressure.

[0014] As a preferred embodiment of the above technical solution, in step 4), in the microdynamic simulation based on the self-consistent coverage model, the surface coverage is first initially estimated. Then, based on the function of coverage and adsorption energy and / or reaction activation barrier obtained in step 3), the adsorption energy and reaction activation barrier under the current coverage environment are obtained. The adsorption energy and reaction activation barrier under the current coverage environment are used as input parameters and substituted into the microdynamic simulation to solve for the rate and actual coverage, and the output coverage is obtained.

[0015] In the above technical solution of this invention, adsorption energy refers to the energy generated during the adsorption process, which can also be understood as the reaction energy of the adsorption process. In the quantitative calculation study of heterogeneous catalysis, adsorption energy calculation refers to the energy generated during the adsorption of adsorbate A on substrate (catalyst) B.

[0016] Adsorption energy E ads The calculation formula is: E ads =E AB -E A -E B

[0017] Among them, E AB E refers to the total energy of system AB after substance A is adsorbed onto substrate B. A E refers to the energy of matter A. B This refers to the energy of the base B.

[0018] Reaction energy refers to the energy change from reactants to products in a one-step elementary reaction. The specific formula is E = E0 FS -E IS

[0019] Here, E refers to the reaction energy of this elementary reaction step. IS This refers to the energy of the initial state of the reaction, E. FS This refers to the energy at the final state of the reaction.

[0020] The activation energy barrier (i.e., activation energy) Ea is the energy difference between the transition state and the initial state in an elementary reaction.

[0021] In the above-mentioned technical solution of this invention, microscopic kinetic simulation refers to the process of first decomposing a complex catalytic system into a series of elementary reactions in the rate study of a heterogeneous catalytic system, then constructing kinetic equations for each elementary reaction based on Arrhenius equations, and finally solving all the kinetic equations simultaneously to obtain the catalytic rate. This process requires the construction of specific elementary reactions for different catalytic systems and the assumption of different degrees of ideal states for the heterogeneous catalytic system.

[0022] In the above-described technical solution of this invention, when performing microscopic kinetic simulation, the input parameters include the reaction energy and activation barrier of each elementary reaction. During the coverage-consistent microscopic kinetic simulation, the actual values ​​of the reaction energy / activation barrier change with the coverage. Therefore, before performing the microscopic kinetic simulation, a preliminary guess of the surface coverage is needed to obtain the reaction energy and activation barrier of each elementary reaction under the current environment.

[0023] As a preferred embodiment of the above technical solution, if the difference between the initial guess coverage and the output coverage is greater than the set convergence criterion, the coverage is guessed again with reference to the output coverage value; if the difference between the initial guess coverage and the output coverage is less than the set criterion, the model is converged, the iteration ends, and the reaction rate under the current free radical pressure is obtained, i.e., the output reaction rate.

[0024] As a preferred embodiment of the above technical solution, after obtaining the output reaction rate, the output reaction rate is compared with the set standard. If the output reaction rate is greater than the set standard, the initial guess of the free radical pressure is made again according to the current situation. If the output reaction rate is less than the set standard, the model converges and the iteration ends. The pressure at this moment is the true partial pressure of the free radical in steady state, and the catalytic reaction rate obtained by microkinetic simulation at this moment is the true steady-state reaction rate, that is, the true steady-state reaction rate is obtained.

[0025] The innovation of this invention lies in first comprehensively considering the reaction mechanism involving free radicals. Simultaneously, a continuous energy change model is established to simulate the influence of the discontinuous "coverage effect" on energy in the molecular sieve system. Finally, by fitting the curve of free radical pressure (used to describe free radical concentration) versus reaction rate, the pressure value under steady-state conditions is obtained, and the reaction rate at this free radical pressure is derived, thus solving the problem of the difficulty in handling free radicals in microscopic kinetic simulations. Based on these innovations, accurate simulation of the dehydrogenation rate of alkanes in molecular sieve systems can be achieved.

[0026] In summary, the present invention has the following beneficial effects:

[0027] 1. This invention accurately predicts the intrinsic dehydrogenation activity of alkane in molecular sieves through microscopic dynamics simulation. It innovates in the model of reaction mechanism, coverage effect, and simulation method of free radical pressure in pores, and realizes the continuous processing of the coverage energy of discontinuous oxygen sites in molecular sieve pores, thereby establishing the functional relationship between coverage and energy. By solving the partial pressure of free radicals in steady state, the fuzzy treatment of free radicals in molecular sieves is solved, and finally the experimental activity data and theoretical data are consistent.

[0028] 2. This invention considers the presence of free radicals in the alkane dehydrogenation reaction mechanism, processes the DFT energy obtained under different coverage conditions at discrete sites in the molecular sieve as a continuous function of coverage and energy change, and determines the free radical pressure in the pores through an iterative method (i.e., a free radical pressure prediction model). The microscopic kinetic simulation performed by this method can accurately predict the intrinsic activity of alkane dehydrogenation in the molecular sieve, achieving a good match between theoretical and experimental results (with an error of less than one order of magnitude). Attached Figure Description

[0029] Figure 1 This invention provides a micro-kinetic simulation-based prediction process for alkane dehydrogenation activity, which includes a free radical pressure prediction model and a coverage self-consistent model.

[0030] Figure 2 This is an approximate model for the continuous change of H* coverage and H* adsorption energy in this invention;

[0031] Figure 3This is an approximate model for the continuous variation of H* coverage and R1 activation barrier in this invention;

[0032] Figure 4 The convergence process of the free radical stress prediction model;

[0033] Figure 5 The theoretical rates calculated under different conditions of the present invention are (i) considering the presence of molecular sieve channels in a single radical molecule, ii) simulating an open system with very low ethyl radical pressure, and iii) representing the results of microdynamic simulations based on radical pressure prediction models and coverage self-consistent models. Detailed Implementation

[0034] The present invention will be further explained and described below with reference to the accompanying drawings.

[0035] This specific embodiment is merely an explanation of the present invention and is not intended to limit the present invention. Any changes made by those skilled in the art after reading the specification of the present invention, as long as they are within the scope of the claims, will be protected by patent law.

[0036] This invention uses Fe / SAPO-34 as a molecular sieve model and ethane dehydrogenation to ethylene as a reaction model for illustrative examples. The specific process is as follows: Figure 1 As shown.

[0037] First, based on the mechanism of alkane dehydrogenation and considering the participation of free radicals, the specific elementary reactions R1-R5 are established as shown below.

[0038] Elementary reaction number Elementary reactions R1 <![CDATA[CH3CH3(g) + * → CH3CH2• + H*]]> R2 <![CDATA[CH3CH2• + * → CH3CH2*]]> R3 <![CDATA[CH3CH2• + * → CH2CH2(g) + H* <!-- 3 -->]]> R4 <![CDATA[CH3CH2* → CH2CH2(g) + H*]]> R5 <![CDATA[2H* → H2(g) + 2*]]>

[0039] Based on the established reaction mechanism, DFT calculations were performed on the activation barriers and adsorption energies of the major adsorbed species for the elementary reactions R1-R5 under different coverage conditions. Here, H* is considered the major covering species, and H* and CH3CH2* are the major adsorbed species. The adsorption energy calculation uses H as an example; the H* adsorption energy equals the difference between the energy of the H* adsorbate and its surface, and the energy of the surface and 0.5 times the hydrogen energy. The energy barrier calculation uses R1 as an example; its energy barrier equals the difference between the transition state energy and the initial state energy at that coverage level, i.e., the maximum energy during the first dehydrogenation bond breaking minus the energy of ethane present on the catalyst surface.

[0040] Since 1 mL is defined as 4 O atoms, 0, 0.25, 0.50, 0.75, and 1 mL correspond to the surface coverage when 0, 1, 2, 3, and 4 adsorbed species are present, respectively. For example... Figure 2 , 3As shown, the energy results calculated for different H* coverage conditions at discrete active sites (i.e., oxygen sites) around the molecular sieve for the adsorbate (or transition state) are continuously processed to construct a function of H* coverage with adsorption energy and activation energy barrier. The coverage value of each point's x-axis is taken as the midpoint of the energy calculated by DFT; for example, the first point is the midpoint between 0 ML and 0.25 ML, i.e., 0.125 ML.

[0041] According to the procedure, an initial estimate of the CH3CH2· free radical pressure is set to 1.02 × 10⁻⁶. -10 bar, such as Figure 4 The ① mark indicates the first iteration of the free radical pressure prediction model. After setting the initial pressure, microscopic dynamic simulations based on the "coverage self-consistent model" can be performed. For example, if the coverage of surface H* is initially assumed to be 0.20 ML, the energy (e.g., the energy when H* coverage is 0.20 ML) can be obtained based on the fitted function of coverage, adsorption energy, and energy barrier. Figure 2 The H* adsorption energy in it is -0.80 eV. Figure 3 The activation barrier of R1 is 1.92 eV. After obtaining all the necessary energy data, we input them into the microdynamic simulation. The actual coverage obtained through the microdynamic simulation with an initial guessed coverage of 0.20 ML is 0.80 ML. This completes the first coverage cycle. Here, the difference between the initial guessed coverage and the output coverage is 0.60 ML, which is greater than the accuracy standard (generally, the difference should be less than 0.01). Therefore, we need to reset the initial guessed coverage and perform microdynamic simulations until the convergence criterion is met. When the initial coverage is set to 0.50 ML, the output coverage is 0.49 ML, which meets the convergence criterion. At this point, the coverage self-consistent model converges, the iteration ends, and the current CH3CH2· free radical pressure is obtained as 1.02 × 10⁻⁶. -10 The kinetic results for bar, H* coverage of 0.50 ML are as follows: Figure 4 Mark point ①. At this point, the reaction rate of the CH3CH2· radical is 3.21 × 10⁻⁶. -1 Because the reaction rate of the CH3CH2· radical is greater than the accuracy convergence value (typically 10), -8 The initial guess of free radical pressure does not meet the convergence criterion of the free radical pressure prediction model. Therefore, the initial guess of free radical pressure needs to be reset until the convergence criterion is met. Figure 4 The iterative processes shown in ①, ②, ③, and ④ are illustrated. When the CH3CH2· free radical pressure is set to 1.24 × 10⁻⁶,... -9 bar, i.e. Figure 4 At point ④, the reaction rate of the CH3CH2· radical is 6.02 × 10⁻⁶. -9 Less than the convergence criterion 10 -8The iteration ends. The pressure at this moment is the true partial pressure of the free radical in steady state, and the catalytic reaction rate obtained through microscopic kinetic simulation is the true steady-state reaction rate, i.e., the CH2CH2 formation rate is 2.38 × 10⁻⁶. -2 .

Claims

1. A kinetic calculation method for estimating the intrinsic activity of alkane dehydrogenation to olefins in molecular sieves, comprising the following steps: 1) Establish a network for catalytic elementary reactions, taking into account the presence of intermediate free radicals; 2) DFT calculation of the adsorption energy of the main species, thereby obtaining the reaction energy of each elementary reaction; at the same time, DFT calculation of the activation energy barrier of each elementary reaction. 3) DFT calculation of adsorption energy and activation energy barrier of each species in molecular sieve at different coverage of major adsorbates at discrete sites; continuous processing of energy at different coverage at discontinuous sites to construct a function of coverage with adsorption energy and / or activation energy barrier; 4) In the free radical pressure prediction model, the free radical pressure is first initially estimated, and then a micro-dynamic simulation including a coverage self-consistent model is performed based on the value of the free radical pressure. In step 4), in the micro-dynamic simulation based on the self-consistent coverage model, the surface coverage is first initially estimated. Then, based on the function of coverage and adsorption energy and / or reaction activation energy barrier obtained in step 3), the adsorption energy and reaction activation energy barrier under the current coverage environment are obtained. The adsorption energy and reaction activation energy barrier under the current coverage environment are used as input parameters and substituted into the micro-dynamic simulation to solve for the rate and actual coverage, and the output coverage is obtained. If the difference between the initial guess coverage and the output coverage is greater than the set convergence criterion, the coverage is guessed again with reference to the output coverage value; if the difference between the initial guess coverage and the output coverage is less than the set criterion, the model has converged, the iteration ends, and the reaction rate under the current free radical pressure is obtained, i.e., the output reaction rate. After obtaining the output reaction rate, compare the output reaction rate with the set standard. If the output reaction rate is greater than the set standard, then make a preliminary guess of the free radical pressure again based on the current situation. If the output reaction rate is less than the set standard, the model has converged and the iteration ends. The pressure at this moment is the true partial pressure of the free radical in steady state, and the catalytic reaction rate obtained by microkinetic simulation at this moment is the true steady-state reaction rate, that is, the true steady-state reaction rate is obtained.

Images