Method for predicting saturation mass concentration of organic compound in organic aerosol

Through ESI FT-ICR MS technology and DBE parameters, a new model was developed to solve the problem of lack of reflecting molecular chemical properties in the existing technology, significantly improve the accuracy of volatile prediction of organic aerosols, and enhance the secondary organic aerosols and the ability to express climate and air quality models.

CN118914399APending Publication Date: 2024-11-08GUANGZHOU INSTITUTE OF GEOCHEMISTRY CHINESE ACADEMY OF SCIENCES
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411055158.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-02
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

Existing element-based models lack the chemical properties of molecular when predicting the volatile nature of organic aerosols, such as unsaturation, resulting in insufficient prediction accuracy, affecting the understanding of secondary organic aerosol formation and atmospheric chemistry.

Method used

Using electrospray ionization Fourier transform ion cyclonic resonance mass spectrometry (ESI FT-ICR MS) analysis technology, combined with equivalent double bond number (DBE) parameters, a new model is developed to accurately predict the saturated mass concentration of organic compounds and improve volatility Prediction accuracy.

Benefits of technology

Significantly improve the accuracy of organic aerosol volatile prediction, enhance understanding of secondary organic aerosol formation and evolution, and improve representation capabilities in climate and air quality models.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118914399A_ABST
    Figure CN118914399A_ABST
Patent Text Reader

Abstract

The invention provides a method for predicting the saturation mass concentration of an organic compound in organic aerosol, which comprises the following steps of: identifying the molecular composition difference of gas and particle phase samples in the atmosphere by using a positive and negative ion electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) technology; intrinsic chemical parameters that determine the volatility of the organic aerosol are determined to predict the volatility of the organic aerosol. The average molecular weight, the carbon mass percent and the equivalent double bond number (DBE) of a particle phase sample are relatively high, but the average oxygen-carbon ratio (O / C) and the oxygen mass percent are relatively low, so that the identified DBE is remarkably and negatively correlated with the volatility based on the result, and the identified DBE is a key parameter. DBE is taken as a new variable to be incorporated into a previous model to improve a model for predicting the volatility of the organic aerosol: Log10C0 = K0-nCbC-nObO-nNbN-nSbS-nDBEbDBE, and the accuracy of the model is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field The invention belongs to the technical field of volatility prediction, and in particular relates to an organic aerosol volatility prediction model and method. Background Art Organic aerosol (OA) is an important component of atmospheric aerosol, which has an important impact on global climate change, atmospheric particulate pollution and human health. OA can be divided into primary organic aerosol (POA) and secondary organic aerosol (SOA) according to its source. OA contains many chemical compounds and occupies most of the total submicron particles in the troposphere (20%-90%). Biogenic and anthropogenic volatile organic compounds (VOCs) are important precursors of OA. Once released into the atmosphere, VOCs undergo gas-phase oxidation under the catalysis of ozone (O3), hydroxyl radicals (OH) or nitrate radicals (NO3) to produce various oxygenated VOCs. These complex processes lead to the generation of secondary organic aerosols (SOA). SOAs are complex mixtures with a wide range of volatility, hygroscopicity, and reactivity. SOAs participate in many chemical reactions and mass transport processes in both the gaseous and particle phases. Additional parameters of volatility, such as O:C and H:C ratios, are important in the two-dimensional framework for the effective representation of SOAs in chemical transport models. The saturated vapor pressure or saturated mass concentration (C0) of a pure compound is an important thermodynamic property that affects the gas-particle partition equilibrium of organic compounds (OC). OCs can be divided into different volatility categories based on their C0 values. These categories include volatile organic compounds (VOCs, C0>10 6.5 μg·m -3 ), medium volatile organic compounds (IVOCs, 10 2.5 μg·m -3 <C0<10 6.5 μg·m -3 ), semi-volatile organic compounds (SVOCs, 10 -0.5 μg·m -3 <C0<10 2.5 μg·m -3 ), low volatile organic compounds (LVOCs, 10 -4.5 μg·m -3 <C0<10 -0.5 μg·m -3 ), very low volatile organic compounds (ELVOCs, 10 -8.5 μg·m -3 <C0<10-4.5 μg·m -3 ) and ultra-low volatile organic compounds (ULVOCs, C0<10 -8.5 μg·m -3 ). Various models have successfully used volatility parameters to calculate the C0 value of OC. The element-based approach is a common method to calculate and predict C0 by utilizing the composition of various elements in a compound, such as carbon (C), nitrogen (N), oxygen (O), and sulfur (S). The widely used element-based model developed by Li et al. can predict the C0 values ​​of OC containing oxygen, nitrogen, and sulfur based on the elemental composition of the OC. This model considers the contribution of atoms such as C, N, O, and S to C0. However, like most other element-based models, it lacks parameters that reflect the chemical properties of the molecules, such as the degree of unsaturation, which indicates the presence of double bonds or rings. Therefore, it is important to identify the key parameters that affect volatility to improve the accuracy of these models and enhance our understanding of the SOA formation process and role in atmospheric chemistry. Understanding the differences in molecular composition between the particle and gas phases is essential to fully elucidate the air-particle partitioning of semivolatile organic compounds (SVOCs). Air-particle partitioning controls the atmospheric fate, wet / dry deposition, long-range transport, and human exposure pathways of SVOCs. Polyurethane foams (PUFs) have been used worldwide for more than 40 years as active and passive samplers to collect a wide range of compounds with moderate to low vapor pressures. Ultra-high resolution electrospray ionization (ESI) Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) has recently been successfully used to characterize the molecular composition of atmospheric aerosols collected on fiber filters, which has improved our understanding of the molecular composition of OCs. The molecular characteristics of complex organics in the gas phase are still poorly understood, which has hindered a comprehensive understanding of the differences in the molecular composition of different OCs in the particle and gas phases, and also hindered the identification of key parameters affecting chemical volatility. In this invention, electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry (ESI FT-ICR MS) was used to determine the molecular composition of gas and particle phase organic carbon (OCs) in urban ambient air samples. Intrinsic chemical parameters that affect the volatility of organic aerosol (OA) were identified, and these parameters were incorporated into a new model to predict volatility in a more accurate way than currently possible; the new model was used to gain insight into the distribution of volatile compounds in the gas and particle phases. These results will improve our understanding of how to incorporate SOA characteristics into models and how to interpret the outputs of these models. These results will also improve our understanding of how to accurately represent the formation and evolution of OA in regional and global climate and air quality models. Summary of the invention In view of the above problems existing in the prior art, the present invention provides a method for predicting the saturated mass concentration of organic compounds in organic aerosols. The method of the present invention obtains a new model for predicting the saturated mass concentration C0 value by considering the structural complexity and unsaturation (DBE) of the compounds. The new model can accurately predict the C0 values ​​of various OC categories, thereby significantly improving the accuracy of predicting the volatility of compounds. In order to achieve the above object, the technical solution adopted by the present invention to solve its technical problem is: A method for predicting the saturation mass concentration of organic compounds in organic aerosols comprises the following steps: (1) Sample collection and extraction: Capture of gaseous and PM using a high-capacity active air sampler equipped with polyurethane foam (PUF) plugs and quartz fiber filters 2.5 The organic compound samples in the phase, polyurethane foam PUF captures a series of medium and low saturated vapor pressure organic compounds, wherein the medium and low saturated vapor pressure organic compounds are semi-volatile-non-volatile organic compounds; The quartz fiber filter paper and the polyurethane foam PUF plug were extracted with dichloromethane by a Soxhlet extractor, respectively. The extracts were concentrated and separated by column chromatography on a silica gel column. The mixture of dichloromethane and acetone was used for elution to obtain gaseous extracts and PM. 2.5 Phase extract. (2) Electrospray ionization tandem Fourier transform ion cyclotron resonance mass spectrometry: The gaseous extract obtained in step (1) and PM 2.5 The phase extracts were analyzed using 9.4T FT-ICR MS in positive and negative ion ESI mode. Molecules that were easily deprotonated were detected in negative ion ESI mode, and molecules that were easily protonated were detected in positive ion ESI mode. A mass range of 150 to 800 Da was used, and 100 consecutive FT-ICR transients were collected, each containing 4 million data points. Peaks with a signal-to-noise ratio greater than 6.0 were analyzed using a mass deviation of ±0.6 ppm. The analyzed molecular formula was expressed as CcHhOoNnSs, where c represents the number of carbon atoms, h represents the number of hydrogen atoms, o represents the number of oxygen atoms, n represents the number of nitrogen atoms, and s represents the number of sulfur atoms. The equivalent double bond number DBE represents the number of double bonds and rings in an organic molecule and is used to evaluate the unsaturation of a compound. DBE is calculated using the following formula: DBE = ch / 2 + n / 2 + 1; The molecules that are easily deprotonated are molecules containing carboxyl, carbazole, sulfonic acid or hydroxyl groups, and the molecules that are easily protonated are molecules containing carbonyl, pyridyl, keto or amine groups; (3) The molecular formula CcHhOoNnSs was classified into organic compounds PUF-only in the gaseous extract, organic compounds present in both the gaseous extract and PM 2.5 Organic compounds PUF & PAR in phase extracts, only in PM 2.5 The organic compounds in the phase extract PAR-only, the parameters of three types of organic compounds were obtained, namely the average molecular mass, the average oxygen mass percentage, the average carbon mass percentage, the average oxygen-carbon ratio and the average DBE. It was found that the average molecular mass, the average oxygen mass percentage, the average carbon mass percentage, the average oxygen-carbon ratio and the average DBE were all related to the volatility of organic matter. The calculation method for the saturated mass concentration of a single organic compound in the organic aerosol was obtained: Log 10 C0=K0–n C b C –n O b O –n N b N –n S b S -n DBE b DBE , where C0 is the saturation mass concentration, n C is the number of carbon atoms, n O is the number of oxygen atoms, n N is the number of nitrogen atoms, n s is the number of sulfur atoms, n DBE is the DBE number, b c , b o , b N , b s , b DBE They are C atoms, O atoms, N atoms, S atoms and DBE to log 10 The contribution of C0 value, K0 is a constant; (4) Substitute the C0 data of 29601 compounds in the Estimation Programs Interface Suite software developed by the U.S. Environmental Protection Agency into the calculation model for least squares fitting to obtain the b values ​​corresponding to each compound. c , b o , b N , b s , b DBE value. (5) Based on the calculation model obtained in step (3) and b obtained in step (4) c , b o , b N , b s , b DBE , the saturated mass concentration C0 of the corresponding organic compound is predicted. Furthermore, for CH compounds: b c =0.291, b DBE =0.283, K0=10.593; For CHO compounds: b c =0.284, b O =0.7, b DBE =0.272, K0=10.106; For CHN compounds: b c =0.389, b N =0.932, b DBE =0.042, K0=9.953; For CHON compounds, b c =0.333, b O =0.658, b N =1.064, b DBE =0.074, K0=8.812; For CHOS compounds, b c =0.177, b O =0.928, b S =0.96, b DBE =0.3, K0=8.87; For CHONS compounds, b c =0.279, b O =1.269, b N =0.893, b S =1.321, b DBE =0.276, K0=10.785. Furthermore, 15 ml of ammonium carbamate was added to the sample before positive ion electrospray ionization FT-ICR MS analysis. Furthermore, the FT-ICR MS was calibrated after use using known oxygen-containing compounds in the sample. Furthermore, the sampling rate of the active air sampler is 1 cubic meter per minute. Furthermore, the polyurethane foam PUF plug has a diameter of 6.2 cm, a height of 7.8 cm, and a density of 0.030 g / cm3. The present invention also provides an application of a model obtained by the method for predicting the saturated mass concentration of organic compounds in organic aerosols, wherein the model is used for climate effect assessment or air pollution control. Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention uses ESI FT-ICR MS to conduct a comprehensive analysis of PUF and particle samples, and the results obtained provide valuable insights into the intrinsic chemical parameters that control OA volatility. Compared with the gas phase, the average molecular weight, carbon mass percentage and equivalent double bond number of organic compounds in the particles are higher, but the average O / C ratio and oxygen mass percentage are lower. The equivalent double bond number is significantly negatively correlated with volatility and is a key indicator. Incorporating the number of DBEs into the new volatility prediction model can improve our understanding and prediction of the gas-particle distribution of volatile compounds. Incorporating DBE into existing models for predicting organic aerosol volatility significantly improves its accuracy. The impact of carbon-carbon double bonds and rings on volatility is huge, especially for hydrocarbons, which cannot be ignored. 2. The prediction results of the present invention for the saturated mass concentration of organic compounds in organic aerosols have important implications for the environment. Determining the most important parameters affecting OA volatility can improve our understanding of the behavior of OA in the atmosphere. This is crucial for predicting the fate of ozone substances, which have a great impact on air quality, climate dynamics and human health. The new volatility prediction model can be used to describe SOA properties in complex SOA models and improve the subsequent interpretation of model outputs to ensure better representation of the formation and evolution of OA in regional and global climate and air quality models. BRIEF DESCRIPTION OF THE DRAWINGS In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Figure 1 (a) The saturated mass concentration (C0) of hydrocarbon (CH) compounds predicted by the model based on the coefficients of our model and the coefficients of the model of Li et al. is compared with the C0 determined by the EPI Suite software in the NCI database. (b) The C0 of a specific hydrocarbon molecule 28 H 18 , C 28 H 30 , C 28 H 42 , C 28 H 50 and C0 of C28H58 predicted by our model and Li's model, and compared with the C0 determined by EPI Suite software from the NCI database. (c) The C0 predicted by our model and Li's model compared with the C0 determined from the PHYSPROP database ( http: / / www.srcinc.com / what-we-do / environmental / scientific- databases.html ) for comparison with the experimental C0 values ​​obtained. Figure 2 .Comparison of predicted C0 values ​​for compounds with different numbers of heteroatoms: The C0 values ​​predicted by our model and Li's model are compared with the C0 values ​​determined by the EPI Suite software from the NCI database. DETAILED DESCRIPTION In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application. The term "comprising" and other equivalent descriptions involved in the specification and claims of this application are intended to cover non-exclusive inclusions, including both the contents that are clearly described in the specification and claims, and the steps or units that are not described in the specification and claims but are inherent in the product, method or structure. A method for predicting the saturation mass concentration of organic compounds in organic aerosols comprises the following steps: (1) Sample collection and extraction Capture of gaseous and PM using a high-capacity active air sampler equipped with polyurethane foam (PUF) plugs and quartz fiber filters 2.5 The organic compound samples in the phase, polyurethane foam PUF captures a series of medium and low saturated vapor pressure organic compounds, wherein the medium and low saturated vapor pressure organic compounds are semi-volatile-non-volatile organic compounds; The quartz fiber filter paper and the polyurethane foam PUF plug were extracted with dichloromethane by a Soxhlet extractor, respectively. The extracts were concentrated and separated by column chromatography on a silica gel column. The mixture of dichloromethane and acetone was used for elution to obtain gaseous extracts and PM. 2.5 Phase extract. Specifically, we selected paired particulate matter and gaseous samples from urban sites in Beijing, Lanzhou, Shijiazhuang, Shanghai, Wuhan, and Guangzhou, which were collected over a continuous 24-h period in the winter of 2018. Each site was equipped with a high-volume active air sampler (Guangzhou Mingya Instruments) and fitted with a PUF plug (6.2 cm in diameter, 7.8 cm in height, density 0.030 g / cm3) and quartz fiber filter paper (Whatman 203 mm × 254 mm) with a sampling rate of 1 m3 / min to capture gaseous and PM 2.5 PUF is able to efficiently accumulate a range of compounds with medium to low vapor pressures. After the samples were processed, the quartz fiber filter paper and PUF plug were extracted with dichloromethane (DCM) by Soxhlet extractor. The extract was concentrated to a small volume and then passed through a multi-layer silica gel column and eluted with a mixture of dichloromethane and acetone to obtain gaseous extracts and PM 2.5 Phase extract. (2) Electrospray ionization tandem Fourier transform ion cyclotron resonance mass spectrometry The extract obtained in step (1) was analyzed using a SolariX XR 9.4T FT-ICR MS system (Bruker Corporation, Billerica, Massachusetts, USA) in positive and negative ion modes. Prior to positive ion electrospray ionization FT-ICR MS analysis, 15 ml of ammonium carbamate was added to the sample to be tested. Molecules that are easily deprotonated (e.g., molecules containing carboxyl, phenolic hydroxyl or hydroxyl groups) can be detected in negative ion mode. Molecules that are easily protonated (e.g., molecules containing basic groups such as carbonyl, keto and amine groups) can be detected in positive ion mode. A mass range of 150 to 800 Da was used, and 100 consecutive FT-ICR transients were collected, each transient containing 4 trillion data points. The MS instrument was calibrated using known alkyl oxide compounds that were abundant in the sample. Peaks with a signal-to-noise ratio greater than 6.0 were analyzed using a mass tolerance of ±0.6 ppm. The molecular formula obtained from the analysis is expressed as CcHhOoNnSs. The double bond equivalent (DBE) represents the number of double bonds and rings in an organic molecule and is used to assess the degree of unsaturation of a compound. DBE can be calculated using the following formula: DBE = ch / 2 + n / 2 + 1. In the negative ion mode, 8141 ± 1202 molecular formulas were detected in the pellet samples with an average molecular weight of approximately 440 Da. Only 3542 ± 875 molecular formulas were detected in the PUF samples with an average molecular weight of approximately 360 Da. The composition of compounds detected in the pellet and PUF samples from all cities was similar. The average number of carbon (C) atoms, average number of hydrogen (H) atoms, double bond equivalent (DBE) value, and modified aromaticity index (AImod) value of the compounds detected in the pellet samples were higher than those in the PUF samples. In contrast, the average number of oxygen (O) atoms and oxygen-carbon ratio (O / C) of the compounds detected in the PUF samples from all cities were lower than those in the pellet samples. Similarly, in the positive ion mode, more molecular formulas were detected in the pellet samples than in the PUF samples. The average number of C atoms, average number of H atoms, DBE value, and AImod value of the compounds detected in the pellet samples were higher, while the average number of O atoms and O / C ratio were lower, similar to the results in the negative ion mode. For the particle and PUF samples, the molecular weight, number of O atoms, and O / C ratio of the compounds detected in the positive ion mode were slightly lower than those in the negative ion mode, but other molecular parameters were similar in the positive and negative ion modes. The different molecular compositions of the compounds detected in the positive and negative ion ionization modes may be due to the different ionization efficiencies of different compounds in different modes. The difference in the molecular composition of OCs detected in the gas and particle phases is crucial to our understanding of atmospheric chemistry as well as the environmental impacts of OCs in ambient air. The mixture of OCs found in the particle phase is more diverse and complex than that in the gas phase. This is probably due to the condensation and aggregation of organic species during aerosol formation, leading to the formation of compounds with larger and more complex molecular structures. The compounds detected in the particle samples consistently had a higher number of C and H atoms and DBE and AImod values ​​than those in the gas samples, indicating that unsaturated and aromatic compounds are more prevalent in the particle samples than in the gas samples. These molecular features are often associated with SOAs formed from compounds emitted during combustion or other reactive sources. The compounds detected in negative ion mode were mainly divided into two categories, one is oxygen-containing (CHO) compounds with one to nine oxygen atoms, and the other is (CHON) compounds containing both oxygen and nitrogen (species with N1O1–N1O10 or N2O1–N2O10). These compound groups each contributed about 50% of the total peak intensity in negative ion mode. The compounds detected in positive ion mode were mainly CHN compounds (N1 and N2), CHO compounds (O1–O6), and CHON compounds (N1O1–N1O8 and N2O1–N2O8). In both positive and negative ion modes, the number of high DBE value compounds detected in particle samples was higher than that in PUF samples. Van Krevelen plots have been successfully applied to FT-ICR MS data before. Van Krevelen plots help interpret the data and clarify the origin and chemical nature of the compounds, as well as the chemical reaction pathways involved in the compounds. The compounds detected in negative ion mode have H / C ratios of 0.4–2.2 and O / C ratios of 0–1. The compounds detected in positive ion mode have a narrower range of O / C ratios (0–0.7). More compounds with H / C ratios of 0.5–1.5 and O / C ratios <0.25 were detected in the particle samples of all cities than in the PUF samples, both in negative and positive ion modes. These compounds are typical SOA components, which are oxygenated derivatives of condensation precursors such as polycyclic aromatic hydrocarbons, and they are formed through complex chemical reactions of organic matter interacting with oxidants in the atmosphere. A large proportion of these compounds have high DBE values ​​and low O / C ratios, which leads to higher average DBE and AImod values ​​and lower O / C ratios for the compounds detected in the particle samples, while the opposite is true for the compounds detected in the PUF samples. (3) The molecular formula CcHhOoNnSs was classified into organic compounds PUF-only in the gaseous extract, organic compounds present in both the gaseous extract and PM 2.5 Organic compounds PUF & PAR in phase extracts, only in PM 2.5 The organic compounds in the phase extract PAR-only, the parameters of three types of organic compounds were obtained, namely the average molecular mass, the average oxygen mass percentage, the average carbon mass percentage, the average oxygen-carbon ratio and the average DBE. It was found that the average molecular mass, the average oxygen mass percentage, the average carbon mass percentage, the average oxygen-carbon ratio and the average DBE were all related to the volatility of organic matter. The calculation method for the saturated mass concentration of a single organic compound in the organic aerosol was obtained: Log 10 C0=K0–n C b C –n O b O –n N b N –n S b S -n DBE b DBE , where C0 is the saturation mass concentration, n C is the number of carbon atoms, n O is the number of oxygen atoms, n N is the number of nitrogen atoms, n s is the number of sulfur atoms, n DBE is the DBE number, b c , b o , b N , b s , b DBEThey are C atoms, O atoms, N atoms, S atoms and DBE to log 10 C0 is the contribution of the value, K0 is a constant. Specifically, the molecular formulas of the particle and PUF samples detected in negative and positive ion modes in different cities were combined to study the molecular composition of the compounds. These molecular formulas were then divided into three groups: molecules detected only in PUF samples (PUF_only), molecules detected in both PUF and particle samples, and molecules detected only in particle samples (PAR_only). Many compounds in the PAR_only group had high DBE values ​​and low H / C ratios (<2) and O / C ratios (<0.3). This indicates that the compounds in the particle phase are more complex and unsaturated than those found only in the gas phase (PUF_only group). There are significant differences in the molecular composition of the compounds associated with the difference in volatility between the PUF_only group and the PAR_only group. As the volatility of the compounds decreases, the proportion of carboxyl-rich lignin / aliphatic molecules (CRAM) molecules decreases from 38% in the PUF_only group to 27% in the PAR_only group. In contrast, the proportions of lipids and unsaturated hydrocarbons increased as the volatility of the compounds decreased, with the proportion of lipids increasing from 13% in the PUF_only group to 40% in the PAR_only group and the proportion of unsaturated hydrocarbons increasing from 4% in the PUF_only group to 20% in the PAR_only group. Lipids, including unsaturated fatty acids, are less volatile and therefore tend to exist in the ambient air in the particle phase. Unsaturated hydrocarbons are usually produced from oxygenated derivatives of polycyclic aromatic hydrocarbons and generally exhibit low volatility. Unsaturated hydrocarbons with heteroatomic groups tend to exist in the particle phase rather than the gas phase. The molecular weight of a compound is closely related to its volatility, with higher molecular weights generally corresponding to lower volatility. The molecular weight of the PUF_only group is lower than that of the PAR_only group, which is consistent with the expectation that volatility decreases with increasing molecular weight. The inverse relationship of the O% value to volatility may initially seem counterintuitive for other reasons, since compounds containing more oxygen atoms are generally thought to be less volatile than compounds containing fewer oxygen atoms. Finally, the significant difference in mean DBE between the PUF_only and PAR_only molecules suggests that DBE is negatively correlated with volatility and may be a key factor affecting volatility. The degree of unsaturation in the compound will undoubtedly affect the behavior of OA. C0 is an important thermodynamic parameter describing the OC gas-particle distribution equilibrium. U., & Shiraiwa, M. (2016). Molecular corridors and parameterizations of volatility in the chemical evolution of organic aerosols. Atmospheric Chemistry and Physics, 16 (5), 3327-3344) established a widely used element-based model to predict C0, which takes into account the contribution of atoms such as C, N, O and S to C0. However, the model does not include key parameters that reflect the chemical properties of OCs, which are essential for accurate prediction of volatility. The present invention aims to improve and optimize the model by introducing the DBE parameter, because DBE was found to strongly affect the volatility of molecules in OA. The updated model is shown below: Log 10 C0=K0–n C b C –n O b O –n N b N –n S b S -n DBE b DBE , where K0 is the reference constant for each compound class; n C 、n O 、n N 、n S and n DBE are the number of C atoms, O atoms, N atoms, S atoms and DBE respectively; b C , b O , b N , b S and b DBE They are C atoms, O atoms, N atoms, S atoms and DBE to log 10 Contribution of C0 value. (4) The C0 data of 29601 compounds in the Estimation Programs Interface Suite software developed by the U.S. Environmental Protection Agency were substituted into the calculation model for least squares fitting to obtain the b values ​​corresponding to different compounds. C , b O , b N , b S and b DBE value. Specifically, b C , b O , b N , bS and b DBE The values ​​of these parameters were determined by performing a multivariate least squares analysis and fitting the model to data from 29,601 compounds of the relevant compound classes (CH, CHO, CHN, CHON, CHOS, and CHONS). Table 1 shows the best fit parameters at 298 K. For CH compounds: b c =0.291, b DBE =0.283, K0=10.593; For CHO compounds: b c =0.284, b O =0.7, b DBE =0.272, K0=10.106; For CHN compounds: b c =0.389, b N =0.932, b DBE =0.042, K0=9.953; For CHON compounds, b c =0.333, b O =0.658, b N =1.064, b DBE =0.074, K0=8.812; For CHOS compounds, b c =0.177, b O =0.928, b S =0.96, b DBE =0.3, K0=8.87; For CHONS compounds, b c =0.279, b O =1.269, b N =0.893, b S =1.321, b DBE =0.276, K0=10.785. We then compared the performance of the proposed method with the results from the Estimation Programs Interface Suite (version 4.1) software developed by the US EPA on data from 29,601 compounds listed by Li et al., where compounds with a C0 value of 0 were excluded.

[0053] Table 1. K0 and b values ​​of saturated mass concentrations of inorganic compounds in the new model

[0054] The correlation coefficient (R) and mean absolute total error (MAGE) were calculated to determine how consistent the results of our method were with the model developed by Li et al. The results (Table 2) show that our model is superior to the model developed by Li et al. in predicting volatility. In particular, for CH, CHO, and CHOS compounds, our model has higher R values ​​and smaller MAGE, indicating that volatility prediction using our model is significantly better than that of Li et al.

[0055] Table 2. Comparative analysis of saturation mass concentration prediction model parameters: our model and Li’s model like Figure 1 As shown in a, the MAGE of our model is 0.27 lower than that of Li's model, representing an improvement of 33.8%. This shows that the impact of carbon-carbon double bonds and rings on volatility is huge, especially for hydrocarbon compounds (CH), and cannot be ignored. MAGE shows that our model improves the accuracy by 7% compared with Li's model. The method developed by Yang et al. (2023) is another element-based model designed to predict the C0 of low-volatility CHO and CHON compounds. We compared the C0 values ​​predicted by our model and Yang's model with the experimental C0 values ​​of 1001 CHO and CHON compounds from the PHYSPROP database. The results show that our method performs well with an R value of 0.8 and a MAGE of 1.42, which is significantly better than Yang's model. These results provide strong evidence that DBE is a key factor affecting the volatility of OA molecules. By considering the structural complexity and unsaturation (DBE) of the compounds, the model is able to accurately predict the C0 values ​​of various OC classes. (5) Based on the calculation model obtained in step (3) and b obtained in step (4) c , b o , b N , b s , b DBE , the saturated mass concentration C0 of the corresponding organic compound is predicted. Specifically, the saturated mass concentration of the corresponding organic compound can be predicted by combining the model and the parameter values ​​corresponding to the different compounds fitted in Table 1. For CH compounds, using the formula Log 10 C0=10.593–0.291n C –0.283n DBE Predict its saturation mass concentration C0; for CHO compounds, use the formula Log 10 C0=10.106–0.284n C–0.7n O –0.272n DBE Predict its saturation mass concentration C0; For CHN compounds, using the formula Log 10 C0=9.953–0.389n C –0.932n N –0.042n DBE Predict its saturation mass concentration C0; For CHON compounds, using the formula Log 10 C0=8.812–0.333n C –0.658n O –1.064n N –0.074n DBE Predict its saturation mass concentration C0; For CHOS compounds, using the formula Log 10 C0=8.87–0.177n C –0.928n O –0.3n DBE Predict its saturation mass concentration C0; For CHONS compounds, using the formula Log 10 C0=10.785–0.279n C –1.269n O –0.893n N –1.321n S –0.276n DBE Predict its saturation mass concentration C0. For CH compounds, C atoms contribute more to C0 than H atoms, but H atoms do make important contributions. The important contribution of H atoms to C0 is the key to our model producing better results than the model developed by Li et al. For CHO compounds, O atoms contribute slightly more to C0 than C atoms. For CHN and CHON compounds, N atoms are the main contributors to C0. For CHOS compounds, the contributions of O and S atoms are similar and greater than the contribution of C atoms to C0. For CHONS compounds, the contributions of N, O, and S atoms to C0 are comparable, but the contribution of O and S is slightly greater than that of N. Heteroatoms including N, O and S have the most significant effect on the volatility of compounds. Heteroatoms have strong electronegativity and can significantly affect the volatility of molecules. Therefore, for compounds containing multiple heteroatoms, the contribution of heteroatoms to volatility is much greater than that of DBE, resulting in only a slight improvement in the performance of our model compared to Li's model. In contrast, for compounds containing a small number of heteroatoms, such as hydrocarbons, the contribution of DBE to volatility is more significant. The prediction results of our model and Li's model for compounds with different numbers of heteroatoms C0 are shown in Figure 2. Figure 2 As shown. For compounds with 0, 1, or 2 heteroatoms, our model shows significant improvement in MAGE compared to Li's model, with improvements of 33.8%, 8.1%, and 10.5%, respectively. However, for compounds with multiple heteroatoms (3 and >3), our model shows limited improvement in MAGE, with improvements of only 3.9% and 2.6%, respectively. Therefore, for compounds with a small number of heteroatoms (0-2), our model significantly improves the accuracy of predicting compound volatility. The improvements provided by our model will be important in a variety of atmospheric and environmental research areas, as accurate volatility predictions are crucial for understanding aerosol formation, behavior, and impacts on air quality and climate. The above-mentioned embodiments are only preferred embodiments of the present invention and cannot be used to limit the scope of protection of the present invention. Any non-substantial changes and substitutions made by technicians in this field on the basis of the present invention shall fall within the scope of protection required by the present invention.

Claims

1. A method for predicting the saturation mass concentration of organic compounds in organic aerosols, characterized in that: The method comprises: (1) Sample collection and extraction: Active air samplers equipped with polyurethane foam (PUF) plugs and quartz fiber filters were used to capture gaseous and PM 2.5 The organic compound samples in the phase, polyurethane foam PUF captures a series of medium and low saturated vapor pressure organic compounds, wherein the medium and low saturated vapor pressure organic compounds are semi-volatile-non-volatile organic compounds; The quartz fiber filter paper and the polyurethane foam PUF plug were extracted with dichloromethane by a Soxhlet extractor, respectively. The extracts were concentrated and separated by column chromatography on a silica gel column. The mixture of dichloromethane and acetone was used for elution to obtain gaseous extracts and PM. 2.5 Phase extract. (2) Electrospray ionization tandem Fourier transform ion cyclotron resonance mass spectrometry: The gaseous extract obtained in step (1) and the PM 2.5 The phase extracts were analyzed using 9.4T FT-ICR MS in positive and negative ion ESI mode. Molecules that were easily deprotonated were detected in negative ion ESI mode, and molecules that were easily protonated were detected in positive ion ESI mode. A mass range of 150 to 800 Da was used, and 100 consecutive FT-ICR transients were collected, each transient containing 4 million data points. Peaks with a signal-to-noise ratio greater than 6.0 were analyzed using a mass deviation of ±0.6 ppm. The analyzed molecular formula was expressed as CcHhOoNnSs, where c represents the number of carbon atoms, h represents the number of hydrogen atoms, o represents the number of oxygen atoms, n represents the number of nitrogen atoms, and s represents the number of sulfur atoms. The equivalent double bond number DBE represents the number of double bonds and rings in an organic molecule and is used to evaluate the unsaturation of a compound. DBE is calculated using the following formula: DBE=ch / 2+n / 2+1. The molecules that are easy to deprotonate are molecules containing carboxyl, carbazole, sulfonic acid or hydroxyl groups, and the molecules that are easy to protonate are molecules containing carbonyl, pyridyl, keto or amine groups; (3) The molecular formula CcHhOoNnSs was classified into organic compounds PUF-only in the gaseous extract, organic compounds present in both the gaseous extract and PM 2.5 Organic compounds PUF & PAR in phase extracts, only in PM 2.5 The organic compounds in the phase extract PAR-only, the parameters of three types of organic compounds were obtained, namely the average molecular mass, the average oxygen mass percentage, the average carbon mass percentage, the average oxygen-carbon ratio and the average DBE. It was found that the average molecular mass, the average oxygen mass percentage, the average carbon mass percentage, the average oxygen-carbon ratio and the average DBE were all related to the volatility of organic matter, and the calculation model of the saturated mass concentration of a single organic compound in the organic aerosol was obtained: Log 10 C0=K0–n C b C –n O b O –n N b N –n S b S -n DBE b DBE , where C0 is the saturation mass concentration, n C is the number of carbon atoms, n O is the number of oxygen atoms, n N is the number of nitrogen atoms, n S is the number of sulfur atoms, n DBE is the DBE number, b c , b o , b N , b s , b DBE They are C atoms, O atoms, N atoms, S atoms and DBE to log 10 The contribution of C0 value, K0 is a constant; (4) Substitute the C0 data of 29601 compounds in the Estimation Programs Interface Suite software developed by the U.S. Environmental Protection Agency into the calculation model for least squares fitting to obtain the b values ​​corresponding to each compound. c , b o , b N , b s , b DBE value; (5) Based on the calculation model obtained in step (3) and b obtained in step (4) c , b o , b N , b s , b DBE , the saturated mass concentration C0 of the corresponding organic compound is predicted.

2. The method according to claim 1, characterized in that In step (4), for CH compound: b c =0.291, b DBE =0.283, K0=10.593; For CHO compounds: b c =0.284, b O =0.7, b DBE =0.272, K0=10.106; For CHN compounds: b c =0.389, b N =0.932, b DBE =0.042, K0=9.953; For CHON compounds, b c =0.333, b O =0.658, b N =1.064, b DBE =0.074, K0=8.812; For CHOS compounds, b c =0.177, b O =0.928, b S =0.96, b DBE =0.3, K0=8.87; For CHONS compounds, b c =0.279, b O =1.269, b N =0.893, b S =1.321, b DBE =0.276, K0=10.

785.

3. The method according to claim 1, characterized in that: Before positive ion electrospray ionization FT-ICR MS analysis, 15 ml of ammonium carbamate was added to the sample.

4. The method according to claim 1, characterized in that: The FT-ICR MS was calibrated after use using known oxygen-containing compounds in the sample.

5. The method according to claim 1, characterized in that The sampling rate of the active air sampler is 1 cubic meter / minute.

6. The method according to claim 1, characterized in that The polyurethane foam PUF plug has a diameter of 6.2 cm, a height of 7.8 cm, and a density of 0.030 g / cm3.

7. An application of a model obtained by the method for predicting the saturated mass concentration of organic compounds in organic aerosols according to any one of claims 1 to 6, characterized in that: The model is used for climate effect assessment or air pollution control.