Fine source apportionment of atmospheric pollutants combining computational chemistry and isotopes

Abstract

The application relates to the field of environmental science and technology, and discloses a fine-source analysis method for atmospheric pollutants combining computational chemistry and isotopes, which comprises the following steps: constructing an isotopic data set, simulating a target pollutant molecular configuration, constructing an isotopic fractionation paradigm equation, correcting local isotopic fractionation, tracing the generation mechanism of a target pollutant in a typical region, and optimizing source analysis of the target pollutant in the typical region. In this way, based on computational chemistry, the density functional theory and the U-B-GM model are coupled, the isotopic fractionation paradigm equation is constructed for a preset reaction path, the specific isotopic fractionation value or fractionation coefficient is accurately simulated in combination with atmospheric environmental parameters during observation, and a local isotopic fractionation correction scheme is established accordingly; the correction scheme is oriented towards fine-source analysis, and can accurately identify the generation mechanism and pollution sources. In this way, the uncertainty of source analysis can be effectively reduced, and fine tracing of the sources of atmospheric pollutants can be realized.

Description

Technical Field

[0001] This application relates to the field of environmental science and technology, for example to a method for refined source analysis of atmospheric pollutants that combines computational chemistry and isotopes. Background Technology

[0002] Atmospheric stable isotope tracing is an important tool for analyzing the sources of air pollutants. This technique uses the stable isotopes of specific elements (such as nitrogen, oxygen, sulfur, and carbon) to trace the sources of air pollutants. 15 N / 14 N、 18 O / 16 O、 34 S / 32 S, 13 C / 12 C) The compositional differences in different pollution sources (such as motor vehicle exhaust, industrial emissions, coal-fired smoke, biomass combustion, etc.) have led to the development of traceable isotope tracing methods.

[0003] Currently, tracing methods based on stable isotopes (quantified by delta) are widely used to reveal the migration and transformation processes of atmospheric pollutants. Taking secondary aerosols as an example, δ... 15 N and δ 18 The O value can be used to determine nitrate (NO3). - The main formation pathways (NO2+OH pathway and N2O5 hydrolysis) and pollution sources (motor vehicle exhaust, industrial emissions, and coal-fired smoke, etc.); utilizing δ 34 The S value can identify sulfate (SO4). 2- The main production mechanisms (H2O2 pathway, OH pathway, and NO2 pathway) and emission sources (coal combustion dust and biomass combustion, etc.) of this technology are affected by isotopic fractionation values ​​(H2O2 pathway, OH pathway, and NO2 pathway). x Due to limitations imposed by α), the isotopic composition of pollutants emitted from the source is not constant. During atmospheric transformation processes (such as photochemical reactions, heterogeneous oxidation, and gas-particle partitioning), kinetic or equilibrium isotopic fractionation occurs, leading to further evolution of the isotopic composition. This evolution of isotopic characteristics caused by isotopic fractionation can mask the isotopic fingerprint information of the original source emissions, resulting in biases in source apportionment results. For example, if NO3 is ignored... - Nitrogen during the formation process ( 15 α) and oxygen ( 18 α) Isotope fractionation effects may overestimate the contribution of coal-fired dust; if SO4 is ignored... 2- Sulfur during the formation process ( 34 α) Isotope fractionation may lead to inaccurate assessments of contributions from sources such as coal combustion dust and biomass combustion. Therefore, the isotope fractionation effect is a key uncertainty in traditional isotope tracing methods that has not yet been fully and effectively corrected.

[0004] Currently, most domestic and international research on atmospheric stable isotope tracing technology focuses on using isotope analysis of typical pollutants in a region (such as NO3). - and SO4 2- The formation mechanisms and sources of pollutants during atmospheric transformation, or the causes of specific pollution events, are still under investigation. However, a systematic and quantitative understanding of the isotopic fractionation mechanisms of typical pollutants during atmospheric transformation is lacking. Existing technologies generally do not provide precise quantification of isotopic fractionation, which directly limits the accuracy and reliability of source apportionment results.

[0005] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this application, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] To provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. This summary is not intended as a general commentary, nor is it intended to identify key / important components or describe the scope of protection of these embodiments, but rather as a prelude to the detailed description that follows.

[0007] This disclosure provides a method for refined source apportionment of atmospheric pollutants by combining computational chemistry and isotopes, in order to reduce the uncertainty of source apportionment caused by fractionation and achieve accurate tracking and emission source tracing of typical atmospheric pollutant reaction pathways.

[0008] In some embodiments, the method for refined source apportionment of atmospheric pollutants combining computational chemistry and isotopes includes: S10, isotope dataset construction: acquiring specific isotopic composition characteristics of the target pollutant, atmospheric environmental parameters during the observation period, and isotopic fractionation reference values ​​to construct an isotope dataset; S20, target pollutant molecular configuration simulation: based on computational chemistry methods, quantitatively simulating and optimizing the molecular geometry of the target pollutant in a specific phase in the atmospheric environment, and obtaining key parameters for isotopic fractionation calculation; S30, isotopic fractionation paradigm equation construction: coupling the computational chemistry method with the UB-GM model to determine isotopic fractionation values; based on the isotopic fractionation reference values, and using nonlinear... A regression algorithm is used to construct the isotopic fractionation paradigm equation for the target pollutant; S40, localized isotopic fractionation correction: based on the atmospheric environmental parameters, the molecular geometry, and the isotopic fractionation paradigm equation, a localized isotopic fractionation correction scheme is constructed; S50, improving the source tracing of the formation mechanism of target pollutants in typical areas: matching the localized isotopic fractionation correction scheme with the isotopic fractionation values ​​of the preset reaction pathway, and using Monte Carlo quantification to quantify the contribution rate of the preset reaction pathway; S60, optimizing the source analysis of target pollutants in typical areas: integrating the contribution rate of the localized isotopic fractionation correction scheme with the preset reaction pathway, constructing and optimizing the source isotopic fingerprint spectrum, and refining the analysis of pollution sources.

[0009] The method for refined source apportionment of atmospheric pollutants combining computational chemistry and isotopes provided in this disclosure can achieve the following technical effects: Based on computational chemistry, this study couples density functional theory (DFT) with the Urey-Bigeleisen-Goeppert-Mayer (UB-GM) model to construct an isotope fractionation paradigm equation for a predefined reaction pathway. Combined with atmospheric environmental parameters (such as temperature, humidity, and pollutant concentrations) during the observation period, it enables the determination of specific isotope fractionation values. x α ) or fractionation coefficient ( x ε This method achieves precise simulation of isotope fractionation and establishes a localized isotope fractionation correction scheme. This correction scheme is used to correlate and optimize the isotope fractionation range of target pollutants in a pre-defined reaction pathway, thereby accurately identifying their formation mechanism and pollution source through refined source apportionment. Based on this concept, this method possesses quantitative correction and qualitative analysis capabilities for isotope fractionation of atmospheric pollutants, effectively reducing the uncertainty of source apportionment and thus achieving refined tracing of primary and secondary sources of atmospheric pollutants.

[0010] The above general description and the description below are exemplary and illustrative only and are not intended to limit this application. Attached Figure Description

[0011] One or more embodiments are illustrated by way of example with reference to the accompanying drawings. These illustrations and drawings do not constitute a limitation on the embodiments. Elements having the same reference numerals in the drawings are shown as similar elements. The drawings are not to be scaled. And wherein: Figure 1 This is a schematic diagram of a method for refined source analysis of atmospheric pollutants combining computational chemistry and isotopes, provided in an embodiment of this disclosure. Figure 2 It is a molecular simulation diagram of atmospheric pollutants based on computational chemistry; Figure 3 These are simulation diagrams based on computational chemistry calculations of isotopes and their key fractionation parameters. Figure 4 This is a schematic diagram of the method for constructing the isotope fractionation paradigm equation provided in the embodiments of this disclosure; Figure 5 This is a verification chart of the isotope fractionation calculation results provided in the embodiments of this disclosure; Figure 6 This is a schematic diagram of the localized isotope fractionation correction method provided in the embodiments of this disclosure; Figure 7 This is a schematic diagram of the source tracing method for improving the formation mechanism of target pollutants in a typical area provided in the embodiments of this disclosure; Figure 8 This is a schematic diagram of the optimized source apportionment method for target pollutants in a typical area provided in the embodiments of this disclosure; Figure 9 This is a diagram illustrating the implementation of isotope tracing technology based on the Bayesian receptor model. Detailed Implementation

[0012] To provide a more detailed understanding of the features and technical content of the embodiments of this disclosure, the implementation of the embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. The accompanying drawings are for illustrative purposes only and are not intended to limit the embodiments of this disclosure. In the following technical description, for ease of explanation, several details are used to provide a full understanding of the disclosed embodiments. However, one or more embodiments may still be implemented without these details. In other cases, well-known structures and devices may be simplified in their depiction to simplify the drawings.

[0013] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this disclosure described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion.

[0014] Unless otherwise stated, the term "multiple" means two or more.

[0015] In this embodiment of the disclosure, the character " / " indicates that the objects before and after it are in an "or" relationship. For example, A / B means: A or B.

[0016] The term "and / or" describes an association between objects, indicating that three relationships can exist. For example, A and / or B means: A or B, or A and B.

[0017] The term "correspondence" can refer to an association or binding relationship. The correspondence between A and B means that there is an association or binding relationship between A and B.

[0018] Combination Figure 1 As shown, this disclosure provides a method for refined source apportionment of atmospheric pollutants combining computational chemistry and isotopes, including: S10, Isotope Dataset Construction: Obtain specific isotopic composition characteristics of the target pollutant, atmospheric environmental parameters during the observation period, and isotopic fractionation reference values ​​to construct an isotope dataset.

[0019] S11, sample and measure the specific isotopic composition characteristics of the target pollutant (e.g., NO3). - δ needs to be tested 15 N and δ 18 O value; SO4 2- δ needs to be tested 34 S and δ 18 O value), used to analyze the localized isotopic composition and fractionation of target pollutants.

[0020] S12, collect the concentration of the target pollutant (e.g., PM2.5). 2.5 and its components, gaseous pollutants such as NO x (e.g., SO2) are used to introduce computational chemistry frameworks to simulate isotope fractionation under specific conditions.

[0021] S13 collects meteorological and geographical data (such as temperature, humidity, wind direction and speed) under the same spatial and temporal conditions, which are then used to introduce a computational chemistry framework to simulate isotope fractionation under specific conditions.

[0022] S14. Collect emission inventories (such as multi-scale emission inventory models) to pre-determine the potential major emission sources of target pollutants in the study area (such as vehicle exhaust, industrial emissions, etc.) as prior information reserves.

[0023] An isotope dataset was constructed using the data collected above.

[0024] S20, Target Pollutant Molecular Configuration Simulation: Based on computational chemistry methods, quantitatively simulate and optimize the molecular geometry of target pollutants in specific phases in the atmospheric environment.

[0025] S21, Identify target pollutants and isotope alternatives: ① Select target pollutant (e.g., NO3) - SO4 2- NO2, SO2, etc. ② Determine isotope substitution schemes (e.g.) 14 N→ 15 N、 16 O→ 18 O、 32 S→ 34 S, 12 C→ 13 (C, etc.) ③ Define the molecular phase (a. gaseous molecules: isolated molecules, without solvent or periodic boundary; b. liquid molecules: solvation effect needs to be considered, such as water, organic solvents, etc.; c. solid molecules: crystal or cluster model or periodic boundary conditions need to be considered).

[0026] S22, Determine the atmospheric environment scenario: ① General scenario construction. Obtain the initial structure of the target pollutant molecule: obtain the initial molecular structure from experimental data or databases; if no experimental data is available, use molecular force fields for preliminary optimization; ② Isotope substitution scenario. Obtaining the isotope substitution structure of the target pollutant molecule: Based on the initial molecular structure of the target pollutant, replace the isotopes of specific atoms (e.g., 14 N→ 15 N), ensuring that the molecular structure is reasonable after replacement (e.g., no significant abrupt changes in bond length or bond angle).

[0027] S23, Selecting the computation method and basis set: ① Quantum chemical methods. Gaseous molecules: DFT (e.g., B3LYP, PBE0) or ab initio calculations (e.g., MP2); Liquid molecules: Implicit solvent models (e.g., PCM) or explicit solvent models (molecular dynamics simulations); Solid molecules: Molecular cluster models or molecular force fields (e.g., COMPASS). ② Basis set selection. Basis set selection for small molecules: 6-31G(d), etc.; Basis set selection for macromolecules: 6-311+G(d,p), etc.; High-precision calculation: such as cc-pVDZ, etc.

[0028] S24, Molecular geometry optimization: Based on optimization algorithms, minimize energy using quasi-Newton methods or conjugate gradient methods until convergence (e.g., energy change < 10). -5Hartree, maximum gradient < 0.001 Hartree / Bohr, displacement change < 0.001 Å).

[0029] S25, Calculate the molecular vibrational frequencies: ① Vibrational frequency calculation. Based on the optimized molecular geometry, calculate the Hessian matrix; use DFT (such as B3LYP / 6-31G(d)) to calculate the vibrational frequencies, and compare the molecular vibrational frequencies in the general scenario with those in the isotope substitution scenario; ② Verify the rationality. Check for imaginary frequencies. If imaginary frequencies are present, it indicates that molecular optimization is not yet complete or the structure is unstable. Ensure that all vibrational modes are positive (i.e., at the potential energy surface minimum).

[0030] S26, Molecular configuration parameter verification and analysis: ① Verify the calculation results. Compare the calculations with experimental data (such as XRD) to verify their accuracy; check for energy convergence to ensure the reliability of the optimization. ② Molecular structure analysis. Optimized bond lengths and bond angles are used to compare the structural differences between general molecules and isotope-substituted molecules; ③ Molecular structure visualization. Use software such as GaussView to draw the optimized molecular configuration, draw vibrational mode diagrams, and label key vibrational frequencies (such as NO stretching and bending).

[0031] Taking computational chemistry simulation driven by GaussView software as an example, the specific steps of step S20 are as follows: ① Determine the target pollutant and its isotope alternatives: Select the target pollutant and its molecular phase for study. In this example, gaseous NO3 is used. - For example; determine the isotope substitution scheme while maintaining gaseous NO3 - With all other atoms in the molecule remaining unchanged, only the nitrogen atom is replaced with an isotope. 14 N→ 15 N); ② Determine the atmospheric environment scenario: based on gaseous NO3 - Molecular and isotopic substitution schemes were used to construct a set of atmospheric environmental scenarios covering both general and isotopic substitution scenarios. The general scenario is gaseous NO3. - The initial molecular structure, based on existing experimental results, is that of gaseous NO3. - The molecule has a tetrahedral structure with a bond angle (∠ONO) of approximately 120° and a bond length (r(NO)) of approximately 1.26 Å. In the isotopic substitution scenario, the mass of the N atom is changed from 14 to 15, while the masses of other atoms remain unchanged. Based on this, gaseous NO3 is constructed. - The initial structure of the molecule and the structure of the isotope substitution; ③ Selection of computational method and basis set. A computational method matching simple gaseous molecules was chosen, using B3LYP for gaseous NO3. - The initial structure of the molecule and the isotopic substitution structure were optimized and the results were converged. ④ Model execution. Combined with... Figure 2 As shown, gaseous NO3 is imported using GaussView software. - The initial molecular structure, isotopic substitution structure, corresponding calculation methods and basis sets of the molecule, and visualization of the initially constructed gaseous NO3. - The molecular model begins iterative calculations to determine the optimal molecular configuration. ⑤ Output results. Combined with... Figure 3 As shown, the necessary input parameters are set using GaussView software, such as the method and basis set in RouteSection, the charge and spin multiplicity of the target molecule in Charge, Multipl., and the molecular structural coordinates of the target molecule in Molecule Specification. Based on this, the optimal molecular geometry of ammonium nitrate molecule under the B3LYP / 6-31G(d) method is run and output, including bond lengths, bond angles, and molecular vibrational frequencies. The software's processing results are used as the basis for verifying whether the vibrational frequency results contain imaginary frequencies and for comparing and validating them with experimentally obtained bond lengths, bond angles, and other parameters.

[0032] In this way, by determining alternatives for the target pollutant and its isotopes, and considering the atmospheric environment scenario, computational chemistry methods and basis sets are selected to optimize the molecular geometry, then the molecular vibrational frequencies are calculated, and finally the optimized molecular geometry is verified and analyzed to obtain an accurate simulation of the target pollutant's molecular geometry.

[0033] Combination Figure 4 As shown in Figure S30, the isotope fractionation paradigm equation is constructed by coupling computational chemistry methods with the UB-GM model to determine isotope fractionation values; based on the isotope fractionation reference values, a nonlinear regression algorithm is used to construct the isotope fractionation paradigm equation for the target pollutant.

[0034] S31, Select a specific reaction pathway, determine the specific reaction pathway mechanism, and clarify the existing forms of reactants and products. This step is used to clarify the basic characteristics of reactants and products to be simulated in computational chemistry.

[0035] Identify specific reaction pathways: ① Select a specific reaction pathway (such as gas-particle partition, photochemical reaction, heterogeneous reaction, etc.); ② Determine the specific reaction pathway mechanism (including parameters such as reactants, products, and reaction rate); ③ Identify the existing forms of reactants and products (including simple existing forms such as gaseous molecules, liquid molecules, and solid molecules, as well as complex mixtures of multiple forms).

[0036] S32, based on computational chemistry methods, simulates the molecular configuration of the target pollutant and compiles the vibrational frequency dataset of the relevant molecules obtained from S25 calculation. This dataset is used to calculate the key parameters of the UB-GM model and then calculate the ratio of the simplified isotope partition function.

[0037] By simulating the molecular configuration of the target pollutant, the vibrational frequencies of the relevant molecules are obtained. Based on the specific reaction pathway mechanism, reactant and product forms determined in S31, a dataset of vibrational frequencies covering reactants and products of the reaction pathway is compiled.

[0038] S33. Set the key parameters of the UB-GM model and calculate the classical factor, zero-point energy contribution, and excitation factor. Use the key parameters, classical factor, zero-point energy contribution, and excitation factor of the UB-GM model as inputs to obtain isotopic fractionation values ​​under different atmospheric environmental parameters.

[0039] Running the UB-GM model: ① Calculation of key parameters. Planck constant. h It is 6.63×10 -34 m²·kg·s -1 Boltzmann constant k 1.38×10 -23 m²·kg·s -2 ·K -1 Speed ​​of light cv i It is 2.99 × 10¹ 0 cm·s -1 ;temperature T Kelvin temperature (K) is used; based on this, it can be determined that... u i = hcv i / kT Calculation based on key parameters u i Values, where the subscripts are... i Indicates the first i A type of molecular vibrational mode.

[0040] ②CF calculation. Based on key parameters. u i Value, based on ∏( u i * / u iCalculate the classical factor CF (translational and rotational energies). Here, "*" represents a molecule substituted with a heavy isotope, while molecules without this superscript correspond to a light isotope.

[0041] ③ZPE calculation. Based on key parameters. u i Value, based on ∏( e -ui* / 2 / e -ui / 2 Calculate the zero-point energy contribution (ZPE). Here, "*" represents molecules substituted by heavy isotopes, while molecules without this superscript are light isotopes. e is a natural constant, representing an exponential function.

[0042] ④ EXC calculation. Based on key parameters. u i Value, according to ∏((1- e -ui ) / (1- e -ui* )) Calculate the excitation factor EXC. Here, "*" represents molecules substituted with heavy isotopes, while molecules without this superscript correspond to light isotopes. e is a natural constant, representing an exponential function.

[0043] ⑤ Model calculation. Key parameters... u i The values, along with the calculation results of CF, ZPE, and EXC, are used as inputs to perform UB-GM model calculations. x β = (CF)(ZPE)(EXC), multiple iterations output different temperatures T. x β value, x β Simplify the distribution function ratio for isotopes; according to x α (product / reactants) = x β product / x β reactants Calculate the values ​​at different temperatures T x α value. x The values ​​represent isotopic abundance, product, and reactants.

[0044] S34 uses a regression algorithm to construct the correlation between different atmospheric environmental parameters and corresponding isotope fractionation values, so that the goodness of fit meets the goodness of fit threshold, and combines experimental results for comparison and verification, forming an isotope fractionation paradigm equation.

[0045] Construction of the isotope fractionation paradigm equation: ① Optimization of nonlinear regression algorithm. Select an appropriate simple regression algorithm (polynomial, exponential, logarithmic, power function, moving average regression, etc.) and correlate different temperatures T with the corresponding... x α The relationship between values ​​affects the goodness of fit. R 2 Meets the requirement of 0.98 or higher; ② Isotope fractionation paradigm equation. The computational paradigm equation obtained by optimization based on the regression algorithm is compared and verified with the measured isotope fractionation results. If the computational results are basically consistent with the experimental results, the constructed isotope fractionation paradigm equation is stored; otherwise, the process returns to step ① for refitting and optimization.

[0046] gaseous HNO3 to particulate NO3 - Taking the reaction pathway involving the transformation of a complex form of existence as an example, the specific steps of step S30 are as follows: ① Identify a specific reaction pathway. Select atmospheric particulate NO3. - The gas-particle partitioning process during formation clarifies particulate NO3. - The main components include solid NO3. - With liquid NO3 - To clarify the main reactant (gaseous HNO3) and product (solid NO3) of this reaction pathway. - and liquid NO3 - ).

[0047] ② Vibrational frequency dataset of relevant molecules. Compilation of data from gaseous HNO3 to particulate NO3. - The gaseous HNO3 and solid NO3 involved in the conversion process - and liquid NO3 - A dataset of molecular vibrational frequencies.

[0048] ③ Model input. Based on key parameters. u i Calculate the values ​​for gaseous HNO3 and solid NO3 respectively. - and liquid NO3 - The CF, ZPE, and EXC values ​​are calculated, and these parameters are input into the UB-GM model for calculation.

[0049] ④ Model Output. The gaseous HNO3 and solid NO3 are calculated using the UB-GM model. - and liquid NO3 - Different temperatures T under normal and isotopic substitution scenarios. x β Value; according to x α(product / reactants) = x β produc t / x β reactants Calculate the values ​​of these pollutants at different temperatures T under different scenarios. x α Values; the gaseous HNO3 was optimized using a nonlinear regression algorithm to obtain the values ​​of solid NO3. - and liquid NO3 - The isotope fractionation paradigm equation for transformation.

[0050] ⑤ Method improvement. The measured gaseous HNO3 was converted to particulate NO3. - The isotopic fractionation values ​​during the conversion process are used as reference values ​​for isotopic fractionation in this reaction process, and are compared and analyzed with the simulation values ​​obtained from computational chemistry. Figure 5 As shown, the simulated values ​​are close to the reference values, indicating that the calculation results of the UB-GM model are consistent with the actual situation. Combining the isotope fractionation paradigm equation obtained in section ④, and based on the isotope mass balance principle, the particulate NO3 was calculated using the measured fractionation reference values. - Medium solid-state NO3 - and liquid NO3 - Proportional reference value f ref Based on the above reference values, an isotopic fractionation paradigm equation for this reaction pathway is constructed. x α ((product / reactants) =Σ f ref(i) x α (producti / reactants) .in, i This indicates the transformation of reactants. i product in class form i ; f ref(i) This indicates that reactants are transformed into products. i The proportional reference value.

[0051] Thus, the UB-GM model is used to calculate isotopic fractionation values ​​for specific reaction pathways, forming an isotopic fractionation paradigm equation. The isotopic fractionation values ​​can be calculated using the simplified isotopic partition function ratios for single-atom exchange. The UB-GM model, constructed by Bigeleisen et al. and Urey et al., considers the classical factor (CF), zero-point energy contribution (ZPE), and excitation factor (EXC), indicating... x β iIt is a function dependent on molecular vibrational frequencies. For specific reaction pathways, the precursors and products before and after the reaction, along with their corresponding molecular vibrational frequencies, must be clearly defined. Nonlinear regression generally employs simple equations, including polynomial, exponential, logarithmic, power function, and moving average regression. Furthermore, the method for calculating isotopic fractionation of target pollutants in complex states within the atmospheric environment has been improved, extending its applicability to isotopic fractionation calculations of mixed states of target pollutants and enhancing the applicability and accuracy of this method under actual atmospheric conditions.

[0052] Combination Figure 6 As shown, S40, Localized Isotope Fractionation Correction: Based on atmospheric environmental parameters, molecular geometry and isotope fractionation paradigm equations, a localized isotope fractionation correction scheme is constructed.

[0053] S41, Determine the dominant formation pathway of the target pollutant: ① Identify the possible reaction pathways of the target pollutant from source emission to formation and transformation. ② Obtain the known isotopic fractionation paradigm equations for the dominant reaction pathway of the target pollutant. ③ Based on the aforementioned possible reaction pathways, the known isotopic fractionation paradigm equations for the dominant reaction pathway, and the molecular geometry, calculate the undefined isotopic fractionation paradigm equations for other reaction pathways of the target pollutant.

[0054] S42, Constructing a traditional isotope fractionation scheme: Using the known isotope fractionation paradigm equations of the target pollutant's dominant reaction pathway, combined with atmospheric environmental parameters during the observation period, calculate the isotope fractionation of the target pollutant's dominant reaction pathway, and formulate a traditional scheme for these existing isotope fractionation studies.

[0055] S43, Optimize the isotope fractionation correction scheme: Based on the established traditional isotope fractionation scheme, incorporate more isotope fractionation paradigm equations that are not yet clear in key reaction pathways, expand isotope fractionation of other reaction pathways that have been neglected in the traditional scheme, and form a more comprehensive and accurate isotope fractionation correction scheme.

[0056] NO3 in atmospheric particulate matter - Taking the isotope fractionation correction scheme of the generation mechanism as an example, the specific operation steps of step S40 are as follows: ① Determine the atmospheric particulate NO3 state - The various formation pathways of NO3 in the atmosphere. Clarify the atmospheric particulate NO3 formation pathways. - Possible reaction pathways during generation and transformation include NO2 + OH, N2O5 hydrolysis, NO3 + VOCs, and N2O5 + Cl. -① Obtain known isotopic fractionation paradigm equations. Through experimental results and literature review, obtain the isotopic fractionation paradigm equations for the NO2+OH pathway and the N2O5 hydrolysis pathway. ② Obtain undefined isotopic fractionation paradigm equations. Combining the molecular configuration of the target pollutant and its isotopic fractionation paradigm equations, calculate other reaction pathways (NO3+VOCs, N2O5+Cl...). - Isotope fractionation paradigm equations for processes such as NO2+OH and N2O5 hydrolysis are developed. ④ A traditional isotope fractionation scheme is constructed. Based on known isotope fractionation paradigm equations and atmospheric environmental parameters during the observation period, such as temperature, the isotope fractionation ranges for the NO2+OH and N2O5 hydrolysis pathways are calculated and compiled into a localized traditional isotope fractionation scheme. ⑤ The isotope fractionation correction scheme is optimized. Based on a more comprehensive isotope fractionation paradigm equation, the calculation of other reaction pathways (NO3+VOCs, N2O5+ClO2+H2O, gas-particle partitioning processes, etc.) is expanded. - The isotope fractionation ranges of processes such as ClONO2+H2O and gas-particle distribution are determined and incorporated into existing traditional schemes to form an isotope fractionation correction scheme.

[0057] This approach integrates the molecular configuration of the target pollutant with its isotopic fractionation paradigm equations, matching the isotopic fractionation calculation methods for the main reaction pathways in its formation and transformation. The localized isotopic fractionation correction scheme emphasizes incorporating more isotopic fractionation data for key reaction pathways that previously lacked quantification, ensuring the comprehensiveness and accuracy of the correction scheme. This breaks away from the traditional approach that only focuses on isotopic fractionation of less quantified reaction pathways, constructing a more comprehensive localized isotopic fractionation correction scheme. The localized isotopic fractionation correction scheme should be calculated in conjunction with specific conditions such as atmospheric environmental parameters in typical regions to better reflect the actual current environmental situation.

[0058] Combination Figure 7 As shown in S50, improve the source tracing of the formation mechanism of target pollutants in typical areas: match the localized isotope fractionation correction scheme with the isotope fractionation value of the preset reaction pathway in S30, and use Monte Carlo to quantify the contribution rate of the preset reaction pathway.

[0059] S51, Construct and configure the model input data file: ① Define the Sources data file. Define the "Sources" data file: Based on prior experience with the formation mechanisms of target pollutants in typical areas, identify possible reaction pathways (taking atmospheric particulate NO3 as an example). - For example, S1: NO2 + OH; S2: N2O5 + H2O; S3: NO3 + VOCs; S4: N2O5 + Cl - S5: ClONO2+H2O, etc.; as particulate SO4 in the atmosphere2- For example, S1: OH; S2: H2O2 / O3; S3: TMI; S4: NO2, etc.), all possible reaction pathways are listed under S. i Enter the ID label column from the Sources file in CSV format. Match the isotope fractionation fingerprint: Based on the isotope fractionation correction scheme built using S40, match the reaction pathway preset in the ID label column of the Sources data file (using S40 as the base). i Import the corresponding isotopic fractionation value (number identifier) ​​to the average value during the observation period. k i The column name for the imported Sources data file is the Mean label column (e.g., Mean15N, Mean18O, Mean34S, etc.). The fractionation values ​​here are corrected using this method, providing a more accurate and complete reflection of the fractionation process in the real atmospheric transformation process. Uncertainty characterization: Standard deviations are matched to the isotopic fractionation values ​​of each reaction pathway to quantify the uncertainty of this parameter. These standard deviations can originate from error analysis of the correction scheme, literature data, or the repeatability of experimental measurements, and are correspondingly matched in the Sources file as the STD label column (e.g., STDd15N, STDd18O, STDd34S, etc.). Generating the Source file: The names (ID column), corresponding optimized isotopic fractionation values ​​(Mean column), and their standard deviations (STD column) of all the above reaction pathways are compiled into a CSV file conforming to the MixSIAR format requirements, serving as the source data input for the model. ② Define the Mixture data file. Integrating Measured Data: Combining the measured isotopic composition of target pollutants in typical regions, and based on the isotopic mass balance principle, the total isotopic fractionation values ​​of the target pollutants from the source to the atmospheric environment are calculated and organized into a data sequence. Generating Mixture Files: The total isotopic fractionation data are organized into a CSV file according to the MixSIAR format requirements, serving as the input for the model's mixture data. This file represents the combined total fractionation process of the unknown generation pathways to be analyzed.

[0060] S52, Setting Prior Probability Distributions: ① Selecting the Prior Type. No-Information Prior: When explicit prior knowledge is lacking, to avoid subjective bias, a no-information prior (such as a uniform distribution) is typically used. This indicates that before the observed data, the contribution rates (0% to 100%) of each response pathway are equally likely, making the final result entirely driven by the measured data. Informative Prior: If expert experience, historical data, or preliminary research results exist, a more informative prior distribution can be set, such as a normal distribution or Dirichlet distribution centered on a specific contribution rate, to guide the model to converge more accurately. ② Configuring Prior Distribution Information. In the MixSIAR user interface, configure the selected prior distribution type and its corresponding parameters for each response pathway.

[0061] S53, Execute Monte Carlo Simulation: ① Input Model Parameters. Load the Source data file and Mixture data file backed up in S51, as well as the preset prior distribution information in S52, into the MixSIAR model. ② Start the MCMC Algorithm. Run the Markov Chain Monte Carlo algorithm in the MixSIAR model. First, perform a diagnostic run, setting the run mode to test to verify the correctness of the model configuration and preliminarily evaluate its performance. After the diagnostic is successful, execute the formal iterative calculations sequentially. Depending on the model complexity and data volume, set the run mode to short, normal, long, very long, or extreme to drive the model to complete tens of thousands to hundreds of thousands of iterations until the Markov chain converges.

[0062] S54, Analysis and Output of Quantitative Results: ① Diagnosing Convergence. After the model runs, tools such as the Gelman-Rubin statistic are used to diagnose whether the MCMC chain has successfully converged, ensuring the reliability of the results. ② Extracting Posterior Probability Distribution. Extracting complete posterior probability distribution samples of the contribution rates of each reaction pathway from the converged Markov chain. ③ Generating Statistical Characterization Results. Performing statistical analysis on the posterior distribution to calculate key statistics that can accurately quantify the contribution rates of each reaction pathway, including: expected value or median, as the best point estimate of the contribution rate of each reaction pathway; 95% confidence interval (or a higher probability interval, such as 75%), used to characterize the uncertainty range of the contribution rate estimate; result visualization and interpretation: outputting the quantitative results in chart form to intuitively show the relative contribution and uncertainty of different preset reaction pathways to the generation of target pollutants. ④ Result Comparison and Evaluation Analysis. The above-mentioned source tracing steps (S51 to S54) were repeated using both the traditional isotope fractionation scheme and the isotope fractionation correction scheme to obtain the corresponding traditional analysis results and the corrected analysis results. By combining quantitative and qualitative methods, the differences in source analysis results under the two schemes were systematically compared and evaluated. The focus was on analyzing the more complete preset reaction pathway introduced in this invention and its impact on the uncertainty of source analysis, thereby verifying and improving the source tracing method of this invention.

[0063] Atmospheric particulate SO4 based on the MixSIAR model 2- Taking the tracing of the generation mechanism as an example, the specific steps of operation S50 are as follows: ① Construct and configure the model input data file. Combined with typical regional atmospheric particulate SO4 2- The main generation pathways were determined, the preset reaction pathways were identified, and their corresponding isotopic fractionation fingerprints and uncertainties were matched and stored in the Source data file in CSV format; combined with the δ-type isotope database constructed by S10, the δ-type isotope was analyzed. 34 SO4 2- value and δ34 SO2 value, based on the principle of isotopic mass balance, according to 34 α total =(δ 34 SO4 2- +1000) / (δ 34 Calculate the atmospheric particulate SO4 (SO2+1000). 2- Total isotope fractionation from the source emission of precursor SO2 to its generation and transformation processes. 34 α total ① Input and store the observed data in the Mixture data file. ② Set the prior probability distribution. To ensure that the implementation avoids subjective bias and that the final result is driven by the measured data, an information-free prior is adopted. ③ Perform Monte Carlo simulation. Import the backed-up Source data file and Mixture data file, along with the preset prior distribution information, into the MixSIAR model; start the Markov chain Monte Carlo simulation algorithm, set run to test, and perform a diagnostic run to verify the correctness of the model configuration; after the diagnosis is successful, perform formal iterative transport, reset the appropriate run mode, and propose a new combination of contribution rates based on the estimated contribution rates of each path and the likelihood function of the model in each iteration. The acceptance-rejection sampling mechanism is used to determine whether to accept the new combination until the Markov chain converges. ④ Model output. Statistical analysis of the posterior distribution is performed, and the statistic of mean ± standard deviation is used to accurately quantify the contribution rate of each reaction pathway. The results are output and quantitative and qualitative analysis of the contribution rate of the generation mechanism is performed. At the same time, the GewekeDiagnostic module is used to perform convergence diagnosis on the Markov chain, calculate the Z-score values ​​of the mean of the initial and tail segments of the chain, and determine the interval in which the Z-score values ​​fall. If the Z-score value falls within the 95% confidence interval, the validity of the posterior distribution can be verified, thereby ensuring the robustness of the model results.

[0064] Thus, the isotope fractionation correction scheme constructed using S40 is applied as a key input to the source tracing model of the target pollutant formation mechanism. This model matches the measured isotopic composition of the pollutant with the isotopic fractionation of each preset reaction pathway, and quantifies the contribution rate of each reaction pathway based on the Monte Carlo simulation method within a Bayesian framework. Specifically, the Monte Carlo simulation constructs an isotopic mass balance model incorporating all preset reaction pathways, combines prior information and measured data, and uses a Markov chain Monte Carlo simulation algorithm for iterative sampling, ultimately outputting the posterior probability distribution of the contribution rate of each pathway, and characterizing the quantification results with expected values ​​and confidence intervals. In particular, this method, by correcting the isotope fractionation scheme, can provide reliable fractionation parameters for potential reaction pathways that were previously excluded due to the difficulty in quantifying isotope fractionation, achieving a more comprehensive and refined analysis of complex formation mechanisms.

[0065] Combination Figure 8 As shown in Figure S60, the source analysis of target pollutants in typical areas is optimized by integrating the contribution rates of local isotope fractionation correction schemes and preset reaction pathways, constructing and optimizing source isotope fingerprint spectra, and refining the analysis of pollution sources.

[0066] S61, Constructing Source Isotope Fingerprints: ① Source Class Identification Guided by Prior Experience. Based on the typical regional isotope database established in step S101, and combined with the emission inventory of the region, systematically identify and classify all potential primary sources of the target pollutant and their secondary precursor source types. Taking atmospheric particulate matter NO3 as an example... - Precursor NO x For example, emissions can be categorized into coal combustion, gasoline vehicle exhaust, diesel vehicle exhaust, biomass combustion, natural gas, and industrial emissions. ② Source isotope data compilation and integration. Isotope value ranges for each source type are obtained through various methods to construct an initial source isotope fingerprint spectrum. Data acquisition methods include: field observation, direct sampling of emission outlets from different sources in typical areas using specific sampling equipment (such as sampling smoke guns and filter membranes, sampling gas bags, etc.), and precise determination using an isotope mass spectrometer to obtain the locally measured isotope composition. Experimental testing: Controlled experiments are conducted on various emission sources (such as combustion of different fuels, specific industrial processes) in laboratories simulating real emission conditions to determine the isotope composition of their products. Literature review: Published literature data similar to the climate, industrial structure, and energy type of the region are systematically reviewed, screened, and organized as supplements and references. Data is compiled by combining the above methods with the actual research scenario. ③ Initial source isotope fingerprint spectrum input and storage. Based on the statistical analysis of the data obtained through the above methods, the mean and standard deviation (STD) of isotopic eigenvalues ​​for each source type are identified and stored in a CSV file in the form of "label: mean ± standard deviation data volume".

[0067] S62, Optimize source isotope fingerprint spectrum: Incorporate the isotope fractionation correction scheme constructed in S40 and the generation mechanism information (such as the relative contribution rate of different reaction pathways) obtained in S50 into the optimization process of source isotope fingerprint spectrum. Specifically, by integrating this prior information, the source isotope values ​​of precursors from the source to the target pollutant in the atmospheric environment are calibrated, including isotope fingerprint correction from precursor to receptor and isotope fingerprint correction from receptor to precursor.

[0068] S63, Refined Source Analysis: ① Data File Preparation. The Source data file contains the optimized source isotope fingerprint spectra (including the isotope average, standard deviation, and sample size for each source), input into the Source data file according to the CSV file format required by the MixSIAR model. The Mixture data file contains the isotopic composition data of the target pollutants actually collected in the environment, input into the Mixture data file according to the CSV file format required by the MixSIAR model. The Discrimination data file can be compiled into a Discrimination data file if additional fractionation factor information exists. In this invention, the fractionation effect has been systematically corrected during the source spectrum optimization stage; therefore, the Discrimination data file can be omitted or its value can be set to 0. ② Model Running and Parameter Setting. In the MixSIAR model's running platform, load the prepared data files (Source data file, Mixture data file, and Discrimination data file); set the key parameters of the model, such as the number of iterations (run, e.g., test, short, normal, long, etc.), to ensure model convergence and obtain stable results. ③ Result Output and Interpretation. Running the MixSIAR model, the final output is the contribution rate of each pollution source ( P k The posterior probability distribution of the model is obtained; the results are diagnosed (such as Gelman-Rubin convergence diagnosis). After confirming that the model has converged, the statistics of the contribution rate of each source (including mean, median, mode and 95% confidence interval) are extracted, so as to achieve a refined and quantitative analysis and uncertainty assessment of pollution sources.

[0069] Atmospheric particulate NO3 based on MixSIAR - Taking the optimization of source resolution methods as an example, the specific operation steps of S60 are as follows: Combined with... Figure 9 As shown, ① the source isotope fingerprint spectrum was constructed and optimized: the MEIC emission inventory based on the S10 isotope database was used to identify the main NO emissions. xEmission sources, including coal combustion (CC), gasoline vehicle exhaust (GV), diesel vehicle exhaust (DV), biomass combustion (BB), natural gas (NG), soil nitrogen emissions (SE), and industrial emissions (IP), were analyzed through literature review. Isotope values ​​for each emission source were matched and stored in the Sources data file as mean ± standard deviation. The constructed isotope fractionation correction scheme was integrated with more comprehensive generation mechanism information obtained from analysis to calculate NO. x δ 15 NO x The values ​​are then imported into the Mixture data file according to the observation sequence. ② The obtained NO values... x The source isotope fingerprint spectrum cannot directly match atmospheric particulate NO3. - The source of NO was determined by analyzing the contribution rates of different reaction pathways and matching them with their corresponding isotopic fractionation values. x To particulate NO3 - Total isotopic fractionation value, based on δ 15 NO3 - =δ 15 NO x + 15 α NO3- / NOx NO, a precursor to the receptor x δ 15 N value. Effectively corrects the source-acceptor isotope signal bias caused by atmospheric chemical transformation processes, reducing the uncertainty in source apportionment. ④ Refine source apportionment: Load the prepared isotope fingerprint spectrum and measured isotope values ​​into the MixSIAR model's runtime platform, start the Markov chain Monte Carlo simulation algorithm, set run to test to perform a diagnostic run first to verify the correctness of the model configuration; after the diagnosis is successful, perform formal iterative transport, reset the appropriate run mode, until the Markov chain converges. ③ Acceptor model output: Perform statistical analysis on the posterior distribution obtained after Markov chain convergence, combined with... Figure 9 As shown, the statistic of mean ± standard deviation is used to accurately quantify the contribution rate of each reaction pathway, outputting the results and performing quantitative and qualitative analysis on the contribution rate output by the refined source analysis. Simultaneously, the Z-score of the chain segment mean is calculated based on the Geweke Diagnostic module. If the Z-score falls within the 95% confidence interval, the sequence is considered convergent, ensuring robustness of the results.

[0070] Thus, the MixSIAR model uses the optimized source isotope fingerprint spectrum as input to the Source data file (in CSV format) and the isotopic composition of the target pollutant samples actually collected in the environment as input to the Mixture data file (in CSV format). It employs a Markov chain Monte Carlo algorithm for random sampling and iterative calculation, ultimately outputting the probability distribution of the contribution rate of each pollution source, thereby achieving a refined and quantitative analysis of pollution sources. The core principle of this model is the construction of a Bayesian hierarchical model, whose mathematical foundation is a mixture equation, i.e. X j =Σ P k ( S jk + c jk )+ ε j ,in X j These are receptor isotope observations from the Mixture data file. P k For the first k The contribution rate of each source to the target pollutant, S k For the first k Isotopes from individual sources j The value of (follows a normal distribution with mean and standard deviation). c jk For the first k Isotopes from individual sources j The fractionation values ​​(follow a normal distribution with mean and standard deviation). ε The residuals (following a normal distribution with a mean of 0 and a standard deviation) are represented by the model contribution rate. P k Treating them as random variables, and using the Markov chain Monte Carlo algorithm to solve their posterior probability distribution, the contributions of each pollution source are accurately quantified and their uncertainty is assessed using statistics such as the average, median, and 95% confidence interval.

[0071] The method for refined source apportionment of atmospheric pollutants, combining computational chemistry and isotopes, provided in this disclosure, is based on computational chemistry and couples density functional theory (DFT) with the Urey-Bigeleisen-Goeppert-Mayer (UB-GM) model. It constructs an isotopic fractionation paradigm equation for a preset reaction pathway and, combined with atmospheric environmental parameters during the observation period (such as temperature, humidity, and pollutant concentration), achieves the apportionment of specific isotopic fractionation values. x α ) or fractionation coefficient ( x εThis method achieves precise simulation of isotope fractionation and establishes a localized isotope fractionation correction scheme. This correction scheme is used to correlate and optimize the isotope fractionation range of target pollutants in a pre-defined reaction pathway, thereby accurately identifying their formation mechanism and pollution source through refined source apportionment. Based on this concept, this method possesses quantitative correction and qualitative analysis capabilities for isotope fractionation of atmospheric pollutants, effectively reducing the uncertainty of source apportionment and thus achieving refined tracing of primary and secondary sources of atmospheric pollutants.

[0072] The technical solutions of this disclosure can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes one or more instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in this disclosure. The aforementioned storage medium can be a non-transitory storage medium, such as a USB flash drive, external hard drive, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, etc., and other media capable of storing program code.

[0073] The foregoing description and accompanying drawings fully illustrate embodiments of this disclosure to enable those skilled in the art to practice them. Other embodiments may include structural, logical, electrical, procedural, and other changes. The embodiments represent only possible variations. Individual components and functions are optional unless explicitly required, and the order of operation may vary. Parts and features of some embodiments may be included or substituted for parts and features of other embodiments. Furthermore, the terminology used herein is for descriptive purposes only and is not intended to limit the claims. Without further limitations, an element defined by the phrase "comprising a…" does not exclude the presence of additional identical elements in the process, method, or apparatus that includes said element. Throughout this document, each embodiment may emphasize differences from other embodiments, and similar or identical parts between embodiments may be referred to mutually. For methods, products, etc., disclosed in the embodiments, if they correspond to the method section disclosed in the embodiments, the relevant details may be referred to the description of the method section.

[0074] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the embodiments of this disclosure. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0075] The methods and products (including but not limited to devices and equipment) disclosed in the embodiments herein can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units may be merely a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to implement this embodiment according to actual needs. In addition, the functional units in the embodiments of this disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0076] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

Claims

1. A method for refined source apportionment of atmospheric pollutants combining computational chemistry and isotopes, characterized in that, include: S10, Isotope Dataset Construction: Obtain specific isotopic composition characteristics of the target pollutant, atmospheric environmental parameters during the observation period, and isotopic fractionation reference values ​​to construct an isotope dataset; S20, Target pollutant molecular configuration simulation: Based on computational chemistry methods, quantitatively simulate and optimize the molecular geometry of the target pollutant in a specific phase in the atmospheric environment; S30, Construction of the isotope fractionation paradigm equation: Determining isotope fractionation values ​​by coupling the computational chemistry method with the UB-GM model; Based on the isotopic fractionation reference values, and using a nonlinear regression algorithm, the isotopic fractionation paradigm equation for the target pollutant is constructed. S40, Localized Isotope Fractionation Correction: Based on the atmospheric environmental parameters, the molecular geometry and the isotope fractionation paradigm equation, a localized isotope fractionation correction scheme is constructed. S50, improve the source tracing of the generation mechanism of target pollutants in typical areas: match the localized isotope fractionation correction scheme with the isotope fractionation value of the preset reaction pathway, and use Monte Carlo to quantify the contribution rate of the preset reaction pathway. S60, Optimize the source analysis of target pollutants in typical areas: Integrate the contribution rates of the localized isotope fractionation correction scheme and the preset reaction pathway, construct and optimize the source isotope fingerprint spectrum, and refine the analysis of pollution sources.

2. The method for refined source apportionment of atmospheric pollutants combining computational chemistry and isotopes according to claim 1, characterized in that, S30 includes: S31, Select a specific reaction pathway, determine the specific reaction pathway mechanism, and clarify the existing forms of reactants and products; S32, Based on computational chemistry methods, by simulating the molecular configuration of the target pollutant, a dataset of vibrational frequencies of relevant molecules is obtained; S33, Based on the vibrational frequency dataset of relevant molecules, set the key parameters of the UB-GM model and calculate the classical factor, zero-point energy contribution and excitation factor; use the key parameters of the UB-GM model, the classical factor, the zero-point energy contribution and the excitation factor as inputs to the UB-GM model to obtain isotopic fractionation values ​​under different atmospheric environmental parameters; S34. Using a regression algorithm, the correlation between different atmospheric environmental parameters and corresponding isotope fractionation values ​​is constructed to ensure that the goodness of fit meets the goodness of fit threshold. The results are then compared and verified with experimental results to form the isotope fractionation paradigm equation.

3. The method for refined source analysis of atmospheric pollutants combining computational chemistry and isotopes according to claim 2, characterized in that, S33 includes: CF=∏( u i * / u i ), where CF is the classical factor; u i These are the key parameters of the UB-GM model; i For the first i A molecular vibrational mode; * indicates a molecule substituted with a heavy isotope; ZPE=∏( e -ui* / 2 / e -ui / 2 ), where ZPE is the zero-point energy contribution and excitation factor; e is a natural constant, representing an exponential function; EXC=∏((1- e -ui ) / (1- e -ui* ), where EXC is the excitation factor; Will u i CF, ZPE, and EXC are used as inputs to import the UB-GM model, and according to... x β = (CF)(ZPE)(EXC) iterates multiple times to output the values ​​at different temperatures T. x β Value, of which, x β Simplify the distribution function ratio for isotopes; according to x α (product / reactants) = x β product / x β reactants Calculate the values ​​at different temperatures T x α Value, of which, x α These are isotopic fractionation values. x The values ​​represent isotopic abundance, product, and reactants.

4. The method for refined source apportionment of atmospheric pollutants combining computational chemistry and isotopes according to claim 1, characterized in that, S40 includes: S41, based on the known isotopic fractionation paradigm equations of the reaction pathways of the target pollutant, and all possible reaction pathways of the target pollutant from source emission to generation and transformation, and in combination with the molecular geometry, calculate the isotopic fractionation paradigm equations of other reaction pathways of the target pollutant that are not yet clear. S42, using the known isotopic fractionation paradigm equation of the dominant reaction pathway of the target pollutant and the atmospheric environmental parameters during the observation period, calculate the isotopic fractionation value of the dominant reaction pathway of the target pollutant, and compile it into a conventional isotopic fractionation scheme. S43. Based on the traditional isotope fractionation scheme, and combined with the isotope fractionation paradigm equation that is not yet clearly defined in the reaction pathway, the isotope fractionation of other reaction pathways that have been neglected in the traditional scheme is expanded to form the isotope fractionation correction scheme.

5. The method for refined source apportionment of atmospheric pollutants combining computational chemistry and isotopes according to claim 4, characterized in that, The S50 includes: S51, define a Sources data file based on the isotope fractionation correction scheme, and define a Mixture data file based on the total isotope fractionation of the target pollutant; S52, Select the prior type and set the prior distribution information; S53, the Sources data file, the Sources data file and the prior distribution information are used as inputs and applied to the MixSIAR model; the Markov chain Monte Carlo algorithm is started in the MixSIAR model; S54, after the MixSIAR model is completed, the posterior probability distribution samples are extracted and analyzed to obtain key statistics on the contribution rate of each response pathway; quantitative and qualitative analysis is performed on the key statistics.

6. The method for refined source apportionment of atmospheric pollutants combining computational chemistry and isotopes according to claim 5, characterized in that, Step S51 includes: Based on prior experience with the formation mechanism of the target pollutants in typical regions, possible reaction pathways are identified, all possible reaction pathways are numbered, and their names are sequentially entered into the ID label column of a Sources data file in CSV format. Based on the isotope fractionation correction scheme, corresponding isotope fractionation values ​​are imported for the reaction pathways preset in the ID label column of the Sources data file. The standard deviation of the isotope fractionation values ​​of each reaction pathway is matched. The ID column of all reaction pathways, the corresponding optimized isotope fractionation values ​​and their standard deviations are organized into a CSV file that conforms to the MixSIAR format requirements, generating the Sources data file as the source data input of the MixSIAR model. Based on the measured isotopic composition of the target pollutant in typical regions and the isotopic mass balance principle, the total isotopic fractionation value of the target pollutant from its source to the atmospheric environment is calculated and organized into a data sequence. The total isotopic fractionation data is then organized into a CSV file according to the MixSIAR format requirements to generate the Mixture data file, which serves as the input of the mixture data information for the MixSIAR model.

7. The method for refined source apportionment of atmospheric pollutants combining computational chemistry and isotopes according to claim 5, characterized in that, Step S54 includes: After the MixSIAR model is completed, diagnose whether the MCMC chain has successfully converged. If so, extract the complete posterior probability distribution sample of the contribution rate of each reaction pathway from the converged Markov chain. Statistical analysis was performed on the posterior probability distribution samples to obtain key statistics on the contribution rate of each response pathway; The conventional isotope fractionation scheme and the corrected isotope fractionation scheme are respectively used to repeat S51 to S54 to obtain the corresponding conventional analysis results and corrected analysis results. The differences in source apportionment results between the two schemes were evaluated by combining quantitative and qualitative methods.

8. The method for refined source apportionment of atmospheric pollutants combining computational chemistry and isotopes according to claim 1, characterized in that, The S60 includes: S61, Statistical analysis is performed on the isotope dataset, the emission inventory of the corresponding region, and the isotope value ranges obtained through multiple methods to classify the source types, to obtain the average value and standard deviation of the isotope characteristic values ​​of each source type, and to form the initial source isotope fingerprint spectrum. S62, using the isotope fractionation correction scheme and the contribution rate of the preset reaction pathway, calibrate the source isotope values ​​of the target pollutant from the source to the atmospheric environment, including isotope fingerprint correction from the precursor to the receptor and isotope fingerprint correction from the receptor to the precursor, so as to obtain the optimized source isotope fingerprint spectrum. S63, Based on the optimized source isotope fingerprint spectrum and the isotopic composition data of the target pollutants in the environment, input data is constructed; the parameters of the analytical model are set, and the input data is input into the analytical model. The analytical model is used to output the posterior probability distribution of the contribution rate of each pollution source, and the statistics of the contribution rate of each pollution source are extracted.

9. The method for refined source apportionment of atmospheric pollutants combining computational chemistry and isotopes according to claim 8, characterized in that, S61 includes: Based on the isotope dataset and the emission inventory of the corresponding region, identify and classify all potential primary sources of the target pollutant and the source types of its secondary precursors; The isotope value ranges for each source type were obtained through multiple methods, and the source isotope data were compiled and integrated; among these methods, the multiple methods included: field observation, experimental testing, and literature review; Based on the statistical analysis of the source isotope data compiled and integrated from the aforementioned source types, the average value and characterization deviation of the isotope characteristic values ​​for each source type are determined and stored in a CSV file in the form of "label: average value ± standard deviation data volume".

10. The method for refined source apportionment of atmospheric pollutants combining computational chemistry and isotopes according to claim 8, characterized in that, S63 includes: The optimized source isotope fingerprint spectrum is input into the Source data file according to the CSV file format required by the MixSIAR model; the isotopic composition data of the target pollutant in the environment is input into the Mixture data file according to the CSV file format required by the MixSIAR model; if there is additional fractionation factor information, it is compiled into a Discrimination data file. In the MixSIAR model's runtime platform, the Source data file, the Mixture data file, and the Discrimination data file are loaded; the key parameters of the MixSIAR model are set. The running MixSIAR model outputs the posterior probability distribution of the contribution rate of each pollution source; the output results are diagnosed, and after confirming that the analytical model has converged, the statistics of the contribution rate of each pollution source are extracted.

Images