A method for identifying HBQ precursors based on FT-ICRMS combined with PMD molecular network
By combining FT-ICRMS with PMD molecular network method, an algal organic matter chlorination reaction system was constructed, the intersection molecular formula was extracted and a mass spectrometry characteristic peak network was constructed, which solved the false positive problem of halobenzoquinone precursor identification in the prior art and achieved efficient and accurate precursor localization.
Patent Information
- Application Number
- CN202611603860.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-10-16
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies struggle to accurately locate precursors of halobenzoquinones in complex water bodies, and the lack of molecular transformation network identification and multidimensional structural parameter joint screening makes it difficult to eliminate false positive interference.
Using FT-ICRMS combined with PMD molecular network method, we established chlorination reaction systems of algae extracellular, intracellular and mixed organic matter, extracted intersection molecular formulas, constructed a mass spectrometry characteristic peak network, and used multidimensional logical screening mechanism and topological parameters to eliminate false positives and locate halobenzoquinone precursors.
This technology effectively extracts target precursors from complex mass spectrometry data, reduces the risk of false positive matches, accurately identifies the formation pathways of halobenzoquinones, and improves the accuracy and reliability of identification.
Smart Images

Figure CN123112295A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of environmental analytical chemistry and water quality testing technology, specifically a method for identifying HBQ precursors based on FT-ICRMS combined with PMD molecular network. Background Technology
[0002] Chlorine disinfection is a common method to ensure drinking water safety. However, the participation of algal organic matter in the reaction can lead to the generation of disinfection byproducts. During the proliferation and death of algae, a large amount of mixed organic matter composed of extracellular and intracellular organic matter is released into the water. During chlorination disinfection, algal organic matter can be converted into toxic disinfection byproducts, including halogenated benzoquinones. The chemical structures of different algal components are significantly different, resulting in different modes of halogenated benzoquinone generation during chlorination. Objectively, it is necessary to identify the molecular characteristics of precursors of halogenated benzoquinones generated by different algal components during chlorination.
[0003] Fourier transform ion cyclotron resonance mass spectrometry (FT-IMS) has become an important tool for characterizing molecular-level changes in disinfection byproducts due to its ultra-high mass resolution and accurate molecular formula identification capabilities. Existing precursor identification methods based on high-resolution mass spectrometry enrich water samples through solid-phase extraction, calculate molecular formulas, extract halogen-containing products, and then match them with samples before disinfection and perform structural classification on Van Krevelen plots.
[0004] Existing precursor identification methods mainly focus on broad classifications such as lignins and proteins, failing to target specific byproduct categories of halobenzoquinones and making it difficult to establish a correlation between precursor molecular formulas and byproduct formation amounts. In data processing, conventional methods often rely on the co-existence of molecular formulas before and after disinfection as a matching criterion, without analyzing the transformation pathways of precursors during chlorination. They lack the ability to utilize molecular networks to reveal the topological relationships of precursor transformations through reactions such as chlorination and decarboxylation. Furthermore, conventional comparison methods lack a joint screening strategy that combines dynamic network transformation characteristics with multi-dimensional structural parameters such as improved aromaticity index, hydrogen-to-carbon ratio, oxygen-to-carbon ratio, and carbon atom number, making it difficult to eliminate false positive matches in non-target mass spectrometry signals. This results in the inability to pinpoint precursor molecules that generate halobenzoquinones from complex aquatic matrices. Summary of the Invention
[0005] To address the shortcomings of existing technologies, this invention provides a method for identifying HBQ precursors based on FT-ICRMS combined with PMD molecular network. This method solves the problem that existing non-targeted mass spectrometry analysis methods lack molecular transformation network identification and multi-dimensional structural parameter joint screening, making it difficult to eliminate false positive interference and accurately locate halogenated benzoquinone precursors from complex water data.
[0006] To achieve the above objectives, the present invention provides the following technical solution:
[0007] This invention provides a method for identifying HBQ precursors based on FT-ICRMS combined with PMD molecular network, employing the following technical solution:
[0008] A method for identifying HBQ precursors based on FT-ICRMS combined with PMD molecular network includes:
[0009] S1. Establish a chlorination reaction system for extracellular organic matter (EOM), intracellular organic matter (IOM), and mixed organic matter (AOM) of algae, and a control sample before chlorination. Sodium hypochlorite was added to the chlorination reaction system, and the reaction was quenched afterward.
[0010] S2, filtration, acidification, and desalting and enrichment via solid-phase extraction;
[0011] S3, UPLC-MS / MS targeted quantification of HBQs and calculation of net generated concentration, FT-ICRMS non-targeted high-resolution determination;
[0012] S4. Molecular formula identification and molecular parameter calculation;
[0013] S5. Extract the molecular formulas that are detected together before and after chlorination in each of the EOM, IOM and AOM systems, and extract the intersection molecular formulas that coexist in the three systems;
[0014] S6. When multiple mass spectrometry characteristic peak records correspond to the same intersection molecular formula, one mass spectrometry characteristic peak record is retained according to the principle of the maximum mass spectrometry response intensity in the three systems of EOM, IOM, and AOM and the six sets of data before and after chlorination. The mass spectrometry characteristic peak retained after deduplication is used as the PMD network node.
[0015] The measured precise quality difference between nodes is matched with the absolute value of the theoretical precise quality difference corresponding to the preset PMD conversion type, and the PMD network is constructed with the conversion relationship corresponding to the quality difference as the connection edge.
[0016] At least one mass spectrometry characteristic peak node that participates in a preset PMD quality difference conversion relationship is a valid node in the PMD network.
[0017] When the percentage of effective nodes participating in the +Cl-H, +Cl-N, or +Cl+OH mass difference conversion is ≥40% of the total number of effective nodes in the PMD network, and the number of connections matching the -CO2 mass difference conversion is ≥3, effective nodes of the PMD network are screened according to AI_mod≥0.5, O / C=0.2~0.6, H / C=0.7~1.5, and the number of carbon atoms is 6~20, and candidate precursor molecular formulas are output.
[0018] By adopting the above technical solution, due to the four-dimensional logical screening mechanism that combines multi-component parallel cross-extraction with precise mass difference molecular network, the effect of extracting target precursor molecules in the background of complex mass spectrometry data is achieved. The algal organic matter system contains a large number of background molecules that are not related to the target reaction. This solution does not directly use data from a single mixed system, but establishes three independent systems: extracellular, intracellular, and mixed.
[0019] By extracting the intersection molecular formulas of the three before and after chlorination, and using the component differences between the systems as cross-validation conditions, the background noise molecules unique to each system are filtered out at the data source.
[0020] Subsequently, node deduplication is performed using the principle of maximizing response strength, providing a basic data source for network computing.
[0021] Based on the node deduplication data, topological mapping is performed by combining the chemical reaction process involved in the formation of halobenzoquinone by environmental organic matter under the action of sodium hypochlorite. Based on the law of conservation of reaction mass, when the precursor molecule undergoes chlorination, the structure loses a hydrogen atom and combines a chlorine atom, resulting in a specific and precise change in mass difference.
[0022] This scheme uses mass spectrometry characteristic peaks as nodes and the measured precise mass difference of the corresponding chemical reaction as the connection edge to reconstruct the molecular transformation network of the chlorination process. The precursor to halobenzoquinone needs to go through a specific reaction path and structural cleavage. This scheme limits the proportion of chlorination-related transformations in the effective nodes of the network to no less than 40% and the number of decarboxylation transformation connections to no less than 3. The topological parameters are used to restrict the target node to be on the transformation path of chlorination reaction and decarboxylation reaction.
[0023] Under the premise that the network topology conditions are met, the physical and chemical property boundaries are superimposed to limit the improved aromaticity index to no less than 0.5, the hydrogen-carbon ratio to be in the unsaturated range of 0.7 to 1.5, and set specific oxygen-carbon ratio and carbon atom number. The physical and chemical logic combination transforms the data network screening into a targeted extraction with chemical structure meaning, and identifies highly aromatic, unsaturated lignin-like precursors, reducing false positive results caused by accidental matching in the network.
[0024] Preferably, in S1, sodium bromide and sodium iodide are added to both the chlorination reaction system and the control sample before chlorination;
[0025] After the reaction is complete, the remaining oxidizing halogen equivalent is measured, and sodium thiosulfate solution is added to quench the reaction, at a rate of 1.5 times the theoretical amount of sodium thiosulfate required for the complete reaction based on the remaining oxidizing halogen equivalent.
[0026] By adopting the above technical solution, bromide ions and iodide ions are introduced into the system to construct a competitive reaction environment in which multiple oxidizing halogens coexist, which meets the halogenation conditions in the water treatment process. Sodium thiosulfate is added according to a specific ratio of the remaining oxidant equivalent to terminate the redox reaction, control the salinity interference caused by excessive quencher, and maintain the ionization efficiency of the system in subsequent mass spectrometry measurements.
[0027] Preferably, in S2, hydrochloric acid is used to adjust the pH to 2;
[0028] A styrene-divinylbenzene polymer solid-phase extraction column was used. After activation with methanol and equilibration with pH=2 hydrochloric acid solution, the sample was loaded and eluted with 3 column volumes of pH=2 hydrochloric acid solution. After drying with nitrogen, the sample was eluted with methanol, concentrated by nitrogen blowing, and then reconstituted with a methanol-water mixture at a volume ratio of 50:50.
[0029] By adopting the above technical solution, the acidity of the system is adjusted to pH 2, so that the organic precursor molecules containing weakly acidic functional groups such as carboxyl groups and phenolic hydroxyl groups are in a non-dissociated neutral state. This enhances the hydrophobic interaction and π-π interaction between the organic molecules and the styrene-divinylbenzene polymer stationary phase, improves the enrichment and recovery rate of the precursors, and, in conjunction with the hydrochloric acid elution step, removes inorganic salt ions from the system, thereby reducing the signal suppression effect of salts on the high-resolution mass spectrometry electrospray ionization process.
[0030] Preferably, in S3, the target HBQs include TCBQ, 2,6-DBBQ and 2,6-DIBQ;
[0031] UPLC-MS / MS employs negative ion multiple reaction monitoring.
[0032] FT-ICRMS employs a negative ion electrospray ionization mode, is calibrated with 10 mmol / L sodium formate and verified with SRNOM, and has a mass error of less than 1 ppm.
[0033] By adopting the above technical solution, quantitative data on the generation of macroscopic products in the system are obtained by using multiple reaction monitoring, and the mass calibration error of high-resolution mass spectrometry is controlled within 1 ppm. This provides physical precision assurance for accurate conversion matching of mass differences and reduces the risk of mismatch in conversion relationship caused by excessively wide error tolerance.
[0034] Preferably, in S4, the elemental composition ranges from C1 to 50, H1 to 100, O1 to 30, NO to 5, SO to 2, and P to 1, with an H / C ratio of 0.3 to 2.5 and an O / C ratio of 0 to 1.2, according to [MH]. - Molecular formula identification was performed using a mass error of 1.0 ppm, a signal-to-noise ratio threshold of 6, and nitrogen rules.
[0035] By adopting the above technical solution, setting elemental composition boundaries and chemometric rules, and using signal-to-noise ratio thresholds and nitrogen rules, random electrical signal fluctuations and distorted molecular formulas that do not conform to the bonding rules of organic elements in the mass spectrum are eliminated, providing a basic molecular library for data processing.
[0036] Preferably, in S4, the double bond equivalent (DBE), the modified aromaticity index (AI_mod), the nominal oxidation state of carbon (NOSC), and the standard Gibbs free energy (GFE) of the organic carbon oxidation half-reaction are calculated based on the identified molecular formula. The calculation formula is as follows:
[0037] ;
[0038] ;
[0039] ;
[0040] ;
[0041] In the formula, when phosphorus is not present in the molecular formula, P takes the value of 0;
[0042] When the denominator parameter C−0.5O−S−N−P equals 0, the corresponding numerator formula is not included in the calculation of AI_mod, nor is it included in the four-dimensional Boolean logic screening that includes the AI_mod constraint.
[0043] By adopting the above technical solution, the basic elemental composition is transformed into a multidimensional structural index that characterizes the degree of molecular unsaturation, aromaticity, and thermodynamic redox state. In the AI_mod calculation, the corresponding molecular node is removed when the denominator is zero. This filters out the interference of carbon-oxygen skeleton-saturated aliphatic chain molecules participating in aromaticity verification from a mathematical logic perspective, thereby improving the rationality of the screening process.
[0044] Preferably, in S6, the preset PMD conversion types include mass difference corresponding to chlorination reaction, mass difference corresponding to +Cl-N type mass difference conversion, mass difference corresponding to chlorination hydroxylation reaction, mass difference corresponding to oxidation reaction, mass difference corresponding to decarboxylation reaction, mass difference corresponding to deoxygenation reaction, and mass difference corresponding to sulfur-containing reaction.
[0045] The measured precise quality difference between nodes is matched with the absolute value of the theoretical precise quality difference corresponding to the preset PMD conversion type, based on a quality difference matching tolerance of 1.0 ppm.
[0046] "+" and "-" indicate the direction of change in the element composition of the corresponding PMD conversion type, and do not indicate the sign of the measured precise quality difference between nodes.
[0047] By adopting the above technical solution, the addition, substitution, oxidation and bond-breaking reaction processes in the halogenation by-product formation pathway are converted into specific mass differences between high-resolution mass spectrometry data. Combined with a tolerance limit of 1.0 ppm, the mutual overlap and interference of different transformation reactions in the mass dimension are reduced, and the chemical reaction mapping relationship is reflected in the topological structure.
[0048] Preferably, in S6, the preset PMD conversion type is matched based on the measured precise mass difference between mass spectrometry characteristic peaks, and the determination of the chlorination mass difference type is not based on whether the molecular formula corresponding to the PMD network node contains the Cl element.
[0049] By adopting the above technical solution, the network structure is broken due to the lack of detection of mass spectrometry signals of low-abundance chlorination products or the lack of molecular formula identification. The network is directly constructed based on the mass difference value, so that the topological association between the parent molecules and structural fragments that have not yet undergone chlorination reaction is established, and the continuity of transformation is maintained.
[0050] Preferably, in S6, for any two PMD network nodes, the difference in measured accurate quality between the nodes is the measured accurate quality of the node with the high mass-to-charge ratio minus the measured accurate quality of the node with the low mass-to-charge ratio.
[0051] By adopting the above technical solution, the direction of the computational scalar of the network connection edge is unified, avoiding confusion in the conversion type recognition caused by the mixed use of positive and negative signs due to poor quality, and maintaining the computational efficiency and matching accuracy of the algorithm when processing batch data.
[0052] Preferably, candidate precursor molecular formulas are classified according to the Van Krevelen region:
[0053] The lipid-like ratio is 1.5 ≤ H / C ≤ 2.2 and 0 ≤ O / C < 0.3;
[0054] The protein / amino acid ratio was 1.5 ≤ H / C ≤ 2.2 and 0.3 ≤ O / C < 0.67.
[0055] The lignin sample has a lignin content of 0.7 ≤ H / C < 1.5 and 0.1 ≤ O / C < 0.67;
[0056] The tannin sample should have a concentration of 0.5 ≤ H / C < 1.5 and a concentration of 0.67 ≤ O / C ≤ 1.2.
[0057] The polycyclic aromatic compounds have a ratio of 0.2 ≤ H / C < 0.7 and 0 ≤ O / C < 0.67.
[0058] By adopting the above technical solution, molecular formulas and node data are mapped back to organic geochemical structural classification intervals, providing structural attribute attribution standards for the screened candidate molecular formulas, and verifying and tracing the characteristics of precursors at the category level.
[0059] This invention provides a method for identifying HBQ precursors based on FT-ICRMS combined with PMD molecular network. It has the following advantages:
[0060] 1. This invention establishes three independent parallel systems: extracellular organic matter, intracellular organic matter, and mixed organic matter. It extracts the intersection molecular formulas detected before and after chlorination of the three systems and retains the unique characteristic peak as a network node based on the principle of maximizing mass spectrometry response intensity. It uses the component differences between different parallel systems for cross-validation and directly filters out the background interference molecules unique to the specific system before data input, avoiding irrelevant environmental background substances from entering the subsequent calculation process and controlling the risk of false positive matching of the initial input data.
[0061] 2. This invention constructs a molecular transformation network by calculating the measured precise mass difference between nodes and matching the measured precise mass difference with the theoretical precise mass difference corresponding to the preset reaction type. It also limits the proportion of nodes participating in chlorination transformation in the network to no less than 40% and the number of decarboxylation connections to no less than 3. It maps the chemical transformation process based on the law of conservation of molecular reaction mass, changes the analysis mode that relies on a single static mass spectrometry signal, and limits the target node to be located in the reaction transformation path that generates halobenzoquinone by relying on topological parameters.
[0062] 3. This invention uses four-dimensional Boolean logic screening in addition to network topology constraints to limit effective nodes to have an improved aromaticity index of not less than 0.5 and to conform to the set range of hydrogen-to-carbon ratio, oxygen-to-carbon ratio and number of carbon atoms. By using physicochemical property boundaries to constrain the structural features of network nodes, it eliminates saturated chain molecules that are accidentally matched in the topology network, and locates lignin-like precursors with high aromaticity and unsaturated characteristics, providing clear molecular structure indicators for the source identification of precursors in complex water environments. Attached Figure Description
[0063] Figure 1 Figure 1 is a schematic diagram of the parameter distribution of candidate precursors under different conditions of the present invention. Figure 2 shows the distribution of the number of molecular formulas and average molecular weight of candidate precursors, and Figure 3 shows the distribution of the average double bond equivalent and average improved aromaticity index of candidate precursors.
[0064] Figure 2 This is a schematic diagram showing the evolution trend of decarboxylation and chlorination conversion nodes in this invention. In Figure (a), the proportion of conversion nodes changes with reaction time, and in Figure (b), the rate of change of the proportion of conversion nodes changes with reaction time.
[0065] Figure 3 This is a schematic diagram of the distribution of results from multiple repeated independent determinations of the present invention. In Figure (a), the distribution of the total number of candidate molecular formulas and the number of common detections is shown, and in Figure (b), the distribution of the average molecular weight of candidate precursors is shown.
[0066] Figure 4 The following is a statistical diagram of the screening results of different data processing methods of the present invention. In the diagram, (a) is a distribution diagram of the total number of detected molecular formulas, and (b) is a diagram of the proportion of molecular formulas with low improved aromaticity index and high hydrogen-to-carbon ratio.
[0067] Figure 5 This is a statistical diagram of network topology parameters under different mass spectrometry measurement accuracies according to the present invention. In the diagram, (a) is a comparison of average mass error and (b) is a diagram of the matching ratio of network connection edges.
[0068] Figure 6 This is a statistical diagram of specific molecules and nodes under different data source processing methods of the present invention. In the diagram, (a) is a distribution diagram of molecular formula and number of nodes, and (b) is a diagram of the proportion of molecular formula and number of nodes.
[0069] Figure 7 Figure 1 is a schematic diagram of the concentration of the generated product and the distribution of the candidate precursors of the present invention. Figure 2(a) is a comparison diagram of the net concentration of halobenzoquinone generated by different organic components, and Figure 3(b) is a projection diagram of the molecular structure region of the candidate precursor. Detailed Implementation
[0070] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0071] The main raw materials and reagents used in the following examples and comparative examples are as follows. Unless otherwise specified, all reagents are commercially available analytical grade or higher grade products.
[0072] Microcystis aeruginosa culture medium uses liquid culture in a stable growth phase as raw material for extracting organic matter from algae.
[0073] Sodium hypochlorite, CAS No. 7681-52-9, is an aqueous solution with an effective chlorine mass fraction of not less than 5%. Before use, its effective chlorine concentration is determined by iodometric titration, and the volume of sodium hypochlorite added to each reaction system is calculated based on the titration results.
[0074] Sodium bromide, CAS number 7647-15-6, analytical grade.
[0075] Sodium iodide, CAS number 7681-82-5, analytical grade.
[0076] Sodium thiosulfate, CAS No. 7772-98-7, analytical grade.
[0077] Hydrochloric acid, CAS number 7647-01-0, is an aqueous solution with a mass fraction of 36% to 38%.
[0078] Sodium hydroxide, CAS number 1310-73-2, analytical grade.
[0079] Methanol, CAS number 67-56-1, chromatographic grade.
[0080] Sodium formate, CAS number 141-53-7, analytical grade.
[0081] A styrene-divinylbenzene polymer solid-phase extraction column, with a matrix of highly cross-linked styrene-divinylbenzene copolymer macroporous resin, is used for desalination and enrichment of dissolved organic matter in water samples.
[0082] Glass fiber filter membranes with pore sizes of 0.7 μm and 0.45 μm, and cellulose acetate filter membranes with a pore size of 0.45 μm.
[0083] Tetrachlorobenzoquinone standard, CAS No.: 118-75-2, purity ≥98%.
[0084] 2,6-Dibromobenzoquinone standard, CAS No.: 19643-45-9, purity ≥98%.
[0085] 2,6-Diiodobenzoquinone standard, purity ≥98%.
[0086] 13C6-2,6-dichloro-1,4-benzoquinone internal standard solution, isotopic purity ≥99%, 100 μg / mL standard solution.
[0087] Acetonitrile, CAS No.: 75-05-8, mass spectrometry grade.
[0088] Formic acid, CAS No.: 64-18-6, mass spectrometry grade, purity ≥99%.
[0089] N,N-Diethyl-p-phenylenediamine reagent, CAS No.: 93-05-0, analytical grade.
[0090] The quality control reference sample for Suwannee River Natural Organic Matter (SRNOM) is from the International Humic Acid Association.
[0091] In one standard preparation method, the culture medium of Microcystis aeruginosa in a stable growth period is centrifuged, and the supernatant is filtered through a filter membrane to obtain extracellular organic matter (EOM).
[0092] The cell pellet obtained by centrifugation was reconstituted with ultrapure water to the original volume, and after repeated freeze-thaw lysis, it was centrifuged again. The supernatant was filtered through a filter membrane to obtain intracellular organic matter (IOM).
[0093] EOM and IOM were mixed in equal volumes to obtain mixed organic matter AOM. The DOC concentration was determined by a total organic carbon analyzer and stored at -20℃ for later use. The specific operating parameters for each preparation example are as recorded in the corresponding preparation example. Algal organic matter can be at least one of EOM, IOM and AOM.
[0094] In the following examples, the EOM, IOM, and AOM systems were set up and measured independently.
[0095] Each preparation example was prepared in parallel scale-up according to the raw material ratio and process conditions recorded in the corresponding preparation example. Before the final dilution and volume adjustment of EOM and IOM, the high DOC mother liquor of EOM and IOM was retained to meet the requirements of the subsequent chlorination reaction and the control sampling volume before chlorination.
[0096] Separately, take the corresponding EOM high DOC mother liquor and mix them with the IOM high DOC mother liquor in equal volumes to obtain the AOM high DOC mother liquor;
[0097] The DOC concentration of each high-DOC mother liquor was determined using a total organic carbon analyzer, and the sampling volume was calculated based on the measured DOC concentration and the target DOC concentration of the corresponding embodiment.
[0098] The DOC concentration of each high-DOC mother liquor was higher than the DOC concentration of the target reaction in the corresponding example.
[0099] The high DOC stock solution is the sample before final dilution and volume adjustment, which is different from the numbered stock solution after final dilution and volume adjustment. In each example, the corresponding high DOC stock solution was adjusted to the target DOC concentration using ultrapure water.
[0100] Preparation Example 1:
[0101] This preparation example provides a method for preparing extracellular organic matter, intracellular organic matter, and mixed organic matter of Microcystis aeruginosa, including the following steps:
[0102] Take 1000 mL of Microcystis aeruginosa culture medium with an absorbance OD680 value of 0.8, centrifuge at 4℃ and 5000 rpm for 10 min, take the supernatant and filter it through a 0.45 μm cellulose acetate membrane, take the corresponding volume of filtrate according to the DOC determination value, and make up to 100 mL with ultrapure water to prepare an extracellular organic matter stock solution with a DOC of 2.0 mg / L, denoted as EOM-1;
[0103] The basal cell pellet obtained by centrifugation was resuspended in 100 mL of ultrapure water. Five freeze-thaw cycles were performed using the repeated freeze-thaw lysis method (each cycle included freezing at -20℃ for 2 h, followed by thawing in a 25℃ water bath for 30 min). The mixture was centrifuged at 4℃ and 5000 rpm for 10 min. The supernatant was filtered through a 0.45 μm cellulose acetate membrane. The corresponding volume of filtrate was taken according to the DOC determination value and diluted to 100 mL with ultrapure water to prepare an intracellular organic matter stock solution with a DOC of 4.0 mg / L, denoted as IOM-1.
[0104] Take 50 mL of EOM-1 stock solution and 50 mL of IOM-1 stock solution respectively and mix them to obtain 100 mL of mixed organic matter stock solution with a DOC concentration of 3.0 mg / L, which is denoted as mixed AOM-1.
[0105] Preparation Example 2:
[0106] This preparation example provides a method for preparing extracellular organic matter, intracellular organic matter, and mixed organic matter of Microcystis aeruginosa, including the following steps:
[0107] Take 1000 mL of Microcystis aeruginosa culture medium with an absorbance OD680 value of 0.8, centrifuge at 4℃ and 4500 rpm for 10 min, take the supernatant and filter it through a 0.45 μm cellulose acetate membrane, take the corresponding volume of filtrate according to the DOC determination value, and make up to 100 mL with ultrapure water to prepare an extracellular organic matter stock solution with a DOC of 3.0 mg / L, denoted as EOM-2;
[0108] The basal cell pellet obtained by centrifugation was resuspended in 100 mL of ultrapure water. Three freeze-thaw cycles were performed using the repeated freeze-thaw lysis method (each cycle included freezing at -20℃ for 2 h, followed by thawing in a 25℃ water bath for 30 min). The mixture was centrifuged at 4℃ and 4500 rpm for 10 min. The supernatant was filtered through a 0.45 μm cellulose acetate membrane. The corresponding volume of filtrate was taken according to the DOC determination value and diluted to 100 mL with ultrapure water to prepare an intracellular organic matter stock solution with a DOC of 3.0 mg / L, denoted as IOM-2.
[0109] Take 50 mL of EOM-2 stock solution and 50 mL of IOM-2 stock solution respectively and mix them to obtain 100 mL of mixed organic matter stock solution with a DOC concentration of 3.0 mg / L, which is denoted as mixed AOM-2.
[0110] Preparation Example 3:
[0111] This preparation example provides a method for preparing extracellular organic matter, intracellular organic matter, and mixed organic matter of Microcystis aeruginosa, including the following steps:
[0112] Take 1000 mL of Microcystis aeruginosa culture medium with an absorbance OD680 value of 0.8, centrifuge at 4℃ and 4000 rpm for 10 min, take the supernatant and filter it through a 0.45 μm cellulose acetate membrane, take the corresponding volume of filtrate according to the DOC determination value, and make up to 100 mL with ultrapure water to prepare an extracellular organic matter stock solution with a DOC of 4.0 mg / L, denoted as EOM-3;
[0113] The basal cell pellet obtained by centrifugation was resuspended in 100 mL of ultrapure water. Three freeze-thaw cycles were performed using the repeated freeze-thaw lysis method (each cycle included freezing at -20℃ for 2 h, followed by thawing in a 25℃ water bath for 30 min). The mixture was centrifuged at 4℃ and 4000 rpm for 10 min. The supernatant was filtered through a 0.45 μm cellulose acetate membrane. The corresponding volume of filtrate was taken according to the DOC determination value and diluted to 100 mL with ultrapure water to prepare an intracellular organic matter stock solution with a DOC of 2.0 mg / L, denoted as IOM-3.
[0114] Take 50 mL of EOM-3 stock solution and 50 mL of IOM-3 stock solution respectively and mix them to obtain 100 mL of mixed organic matter stock solution with a DOC concentration of 3.0 mg / L, which is denoted as mixed AOM-3.
[0115] Preparation Example 4:
[0116] This preparation example provides a method for preparing extracellular organic matter, intracellular organic matter, and mixed organic matter of Microcystis aeruginosa, including the following steps:
[0117] Take 1000 mL of Microcystis aeruginosa culture medium with an absorbance OD680 value of 0.8, centrifuge at 4℃ and 4500 rpm for 10 min, take the supernatant and filter it through a 0.45 μm cellulose acetate membrane, take the corresponding volume of filtrate according to the DOC determination value, and make up to 100 mL with ultrapure water to prepare an extracellular organic matter stock solution with a DOC of 2.5 mg / L, denoted as EOM-4;
[0118] The basal cell pellet obtained by centrifugation was resuspended in 100 mL of ultrapure water. Three freeze-thaw cycles were performed using the repeated freeze-thaw lysis method (each cycle included freezing at -20℃ for 2 h, followed by thawing in a 25℃ water bath for 30 min). The mixture was centrifuged at 4℃ and 4500 rpm for 10 min. The supernatant was filtered through a 0.45 μm cellulose acetate membrane. The corresponding volume of filtrate was taken according to the DOC determination value and diluted to 100 mL with ultrapure water to prepare an intracellular organic matter stock solution with a DOC of 3.5 mg / L, denoted as IOM-4.
[0119] Take 50 mL of EOM-4 stock solution and 50 mL of IOM-4 stock solution respectively and mix them to obtain 100 mL of mixed organic matter stock solution with a DOC concentration of 3.0 mg / L, which is denoted as mixed AOM-4.
[0120] Preparation Example 5:
[0121] This preparation example provides a method for preparing extracellular organic matter, intracellular organic matter, and mixed organic matter of Microcystis aeruginosa, including the following steps:
[0122] Take 1000 mL of Microcystis aeruginosa culture medium with an absorbance OD680 value of 0.8, centrifuge at 4℃ and 4500 rpm for 10 min, take the supernatant and filter it through a 0.45 μm cellulose acetate membrane, take the corresponding volume of filtrate according to the DOC determination value, and make up to 100 mL with ultrapure water to prepare an extracellular organic matter stock solution with a DOC of 3.5 mg / L, denoted as EOM-5;
[0123] The basal cell pellet obtained by centrifugation was resuspended in 100 mL of ultrapure water. Three freeze-thaw cycles were performed using the repeated freeze-thaw lysis method (each cycle included freezing at -20℃ for 2 h, followed by thawing in a 25℃ water bath for 30 min). The mixture was centrifuged at 4℃ and 4500 rpm for 10 min. The supernatant was filtered through a 0.45 μm cellulose acetate membrane. The corresponding volume of filtrate was taken according to the DOC determination value and diluted to 100 mL with ultrapure water to prepare an intracellular organic matter stock solution with a DOC of 2.5 mg / L, denoted as IOM-5.
[0124] Take 50 mL of EOM-5 stock solution and 50 mL of IOM-5 stock solution respectively and mix them to obtain 100 mL of mixed organic matter stock solution with a DOC concentration of 3.0 mg / L, which is denoted as mixed AOM-5.
[0125] In the following examples, comparative examples, and test cases, when calculating molecular formula parameters using high-resolution mass spectrometry, a unified data processing rule was adopted. After removing isotope peaks from the high-resolution mass spectrometry data, molecular formula identification and reasonableness limits were performed according to unified parameters. Except as otherwise specified in Comparative Example 3, the molecular formula identification mass error was set to 1.0 ppm, the signal-to-noise ratio threshold was set to 6, and the allowed elemental composition range was C1–50, H1–100, O1–30, NO–5, SO–2, and P0–1. The ionic form was based on [MH]. - Identification was performed, and the limits were set according to the H / C atomic ratio of 0.3 to 2.5, the O / C atomic ratio of 0 to 1.2, and the nitrogen rule. When the same mass spectrum peak corresponds to multiple candidate molecular formulas, the molecular formula with the fewest heteroatoms was selected first. When multiple candidate molecular formulas still exist, the molecular formula with the smallest mass error was selected.
[0126] Calculate the double bond equivalent (DBE), modified aromaticity index (AI_mod), oxygen / carbon ratio (O / C), hydrogen / carbon ratio (H / C), number of carbon atoms, nominal oxidation state of carbon (NOSC), and standard Gibbs free energy (GFE) of the organic carbon oxidation half-reaction (kJ / molC) based on the identified molecular formula.
[0127] ;
[0128] ;
[0129] ;
[0130] .
[0131] In the formula, C, H, O, N, S and P represent the number of carbon, hydrogen, oxygen, nitrogen, sulfur and phosphorus atoms in the corresponding molecular formula, respectively; when there is no phosphorus element in the molecular formula, P is 0; in this embodiment, the molecular formulas participating in the screening of candidate precursors are all calculated according to the above rules.
[0132] In this embodiment, when C−0.5O−S−N−P=0, the molecular formula is not included in the calculation of AI_mod and AI_mod_w, and is not included in the four-dimensional Boolean logic screening that includes the AI_mod constraint.
[0133] Using the mass spectral peak intensities corresponding to each molecular formula as weights, the weighted average values of the mass spectral peak intensities of H / C_w, O / C_w, DBE_w, GFE_w, and AI_mod_w are calculated respectively. AI_mod_w is calculated by weighting the molecular formulas with valid AI_mod values and their corresponding mass spectral peak intensities, and the sum of the mass spectral peak intensities corresponding to the molecular formulas participating in the AI_mod_w calculation is used as the weighted normalized denominator.
[0134] The 15 sets of data from each reaction system of EOM, IOM and AOM in Examples 1 to 5 constitute a cross-parameter combination correlation analysis dataset. The pre-chlorination parameters of each system were paired with the corresponding net concentrations of TCBQ, 2,6-DBBQ and 2,6-DIBQ for Spearman rank correlation analysis. This analysis only characterizes the statistical correlation between the 15 sets of cross-parameter combination data.
[0135] Correlation was characterized by Spearman's rank correlation coefficient ρ, and p < 0.05 was used as the criterion for statistical significance. The ρ values for H / C_w and the net formation concentrations of TCBQ, 2,6-DBBQ and 2,6-DIBQ were 0.76, 0.68 and 0.82, respectively, with corresponding p values of 0.002, 0.008 and 0.001, respectively. All ρ values were > 0 and all p values were < 0.05. The ρ and p values for the other indicators were calculated in the same way.
[0136] Furthermore, molecular formulas that meet the aforementioned identification criteria are divided according to the Van Krevelen region, wherein:
[0137] The range of lipid-like molecular formulas is 1.5 ≤ H / C ≤ 2.2 and 0 ≤ O / C < 0.3;
[0138] The protein / amino acid-like molecular formula is 1.5 ≤ H / C ≤ 2.2 and 0.3 ≤ O / C < 0.67;
[0139] The molecular formula of the lignin sample is 0.7≤H / C<1.5 and 0.1≤O / C<0.67;
[0140] The molecular formula of the tannin sample is 0.5≤H / C<1.5 and 0.67≤O / C≤1.2;
[0141] The molecular formula of the fused-ring aromatic type is 0.2≤H / C<0.7 and 0≤O / C<0.67.
[0142] Unless otherwise explicitly stated, the calculation of the above molecular parameters in the following examples, comparative examples and test cases shall all follow the calculation rules in this section.
[0143] Example 1:
[0144] This embodiment provides a method for identifying halogenated benzoquinones (HBQs) precursors during the chlorination disinfection process of algal organic matter based on high-resolution mass spectrometry, including the following steps:
[0145] S1. Take the required volume of EOM high DOC mother liquor, IOM high DOC mother liquor and AOM high DOC mother liquor corresponding to Preparation Example 1 according to the target DOC concentration, and establish parallel chlorination reaction systems independently. Take the corresponding EOM high DOC mother liquor, IOM high DOC mother liquor and AOM high DOC mother liquor from the same batch as control samples before chlorination.
[0146] Ultrapure water was used to bring the volume to 100 mL to ensure that the DOC concentration of the above reaction systems and control samples was 5.0 mg / L. Sodium bromide and sodium iodide were added to each chlorination reaction system and the corresponding pre-chlorination control sample, respectively, to ensure that the Br content was within acceptable limits. - Concentration of 500 μg / L, I - The concentration was 500 μg / L, and the initial pH was adjusted to 9.0 using sodium hydroxide or hydrochloric acid;
[0147] Sodium hypochlorite was added to each chlorination reaction system to make the effective chlorine (calculated as Cl2) concentration 25 mg / L. No sodium hypochlorite was added to the control sample before chlorination.
[0148] Each chlorination reaction system was reacted at a constant temperature of 25°C for 48 hours under light-protected conditions. After the reaction was completed, the reaction solution was measured by N,N-diethyl-p-phenylenediamine spectrophotometry to obtain the remaining oxidizing halogen equivalent (calculated as Cl2). Based on the measurement results, the remaining oxidizing halogen equivalent was calculated, and sodium thiosulfate solution was added at 1.5 times the theoretical amount of sodium thiosulfate required for complete reaction to quench the reaction.
[0149] S2. The control sample before chlorination and the three reaction solutions after quenching were pre-filtered through a 0.7μm glass fiber membrane and a 0.45μm glass fiber membrane, respectively, and then filtered through a 0.45μm cellulose acetate membrane. The pH of the samples was adjusted to 2 using hydrochloric acid.
[0150] Subsequently, desalting and enrichment were performed using a styrene-divinylbenzene polymer solid-phase extraction column. Before use, the solid-phase extraction column was activated with methanol and equilibrated with hydrochloric acid solution at pH=2. After the sample was loaded onto the column, it was eluted with 3 column volumes of hydrochloric acid solution at pH=2 to remove inorganic salts. After drying with nitrogen, it was eluted with 20 mL of chromatographic grade methanol. The eluent was concentrated by nitrogen blowing and then reconstituted with a methanol-water mixture at a volume ratio of 50:50 and brought to a final volume of 1.0 mL to obtain the samples to be tested.
[0151] S3. Ultra-high performance liquid chromatography-tandem mass spectrometry (UPLC-MS / MS) in negative ion mode using multiple reaction monitoring (MRM) was employed for targeted quantitative analysis of TCBQ, 2,6-DBBQ, and 2,6-DIBQ in the control sample before chlorination and the sample after chlorination. The concentrations of the samples before and after chlorination were determined, and the net generated concentration was calculated (net generated concentration = concentration after chlorination - concentration before chlorination). Specifically, a Waters ACQUITY UPLC-BEHC18 column (2.1 × 100 mm, 1.7 μm) was used for separation. Mobile phase A was 0.1% formic acid aqueous solution, and mobile phase B was acetonitrile. The gradient elution program was as follows:
[0152] 0–2.0 min, 10% B;
[0153] From 2.0 to 8.0 minutes, the proportion of phase B increased uniformly from 10% to 90%.
[0154] 8.0–10.0 min, 90% B;
[0155] Between 10.0 and 10.1 minutes, the proportion of phase B decreased from 90% to 10%.
[0156] 10.1–13.0 min, 10% B.
[0157] The flow rate was 0.3 mL / min, the column temperature was 40 °C, and the injection volume was 5.0 μL.
[0158] Among them, the MRM quantitative ion pairs (m / z, parent ion / daughter ion) of TCBQ, 2,6-DBBQ and 2,6-DIBQ are 245.8 / 210.8, 264.8 / 236.8 and 358.8 / 330.8, respectively, with corresponding collision energies of 15 eV, 20 eV and 20 eV, respectively;
[0159] TCBQ, 2,6-DBBQ, and 2,6-DIBQ were quantified using the internal standard method. 13 C6-2,6-dichloro-1,4-benzoquinone was used as the internal standard, and the final concentration of the internal standard in the standard curve series and the test samples was 10.0 μg / L.
[0160] The MRM ion pair of the internal standard is 181.0 / 145.0 m / z, corresponding to a collision energy of 15 eV;
[0161] The corresponding standard curve concentration ranges are 0.1–50.0 μg / L, 0.1–50.0 μg / L, and 0.1–50.0 μg / L, respectively. The sample concentration is calculated based on the corresponding standard curve.
[0162] Simultaneously, Fourier transform ion cyclotron resonance mass spectrometry (FT-ICRMS) was used in negative ion electrospray ionization mode to perform non-targeted high-resolution measurements on the control sample before chlorination and the sample after chlorination. The mass spectrometer parameters were set as follows:
[0163] The continuous injection flow rate was 150 μL / h, the capillary inlet voltage was 4.5 kV, the ion accumulation time was 0.1 s, the m / z acquisition range was 100–1600, and the number of time-domain signal superpositions was 400.
[0164] Mass calibration was performed using 10 mmol / L sodium formate, and SRNOM was used as a quality control reference sample for verification to ensure that the mass error was less than 1 ppm, thus obtaining the raw high-resolution mass spectrometry data for each group.
[0165] S4. According to the above unified data processing rules, perform isotope peak removal, molecular formula identification and molecular parameter calculation on each group of high-resolution mass spectrometry raw data to obtain DBE, AI_mod, O / C, H / C, number of carbon atoms, NOSC and GFE for each molecular formula, and calculate H / C_w, O / C_w, DBE_w, GFE_w and AI_mod_w.
[0166] The weighted indices and corresponding net concentrations of TCBQ, 2,6-DBBQ and 2,6-DIBQ of the pre-chlorination control sample were included in the above unified Spearman rank correlation analysis.
[0167] Spearman rank correlation analysis was used to statistically analyze the correlation between system-level molecular parameters and net HBQs generation concentration. The results of this analysis were not used as screening criteria for individual candidate precursor molecular formulas in steps S5 and S6.
[0168] Based on the ρ and p values obtained from the unified Spearman rank correlation analysis above, record the correlation direction and significance between H / C_w, O / C_w, DBE_w, GFE_w, and AI_mod_w and the net generation concentrations of the three HBQs.
[0169] S5. First, compare the mass spectrometry results of the control samples before chlorination and the samples after chlorination of the EOM, IOM and AOM reaction systems corresponding to Preparation Example 1, and extract the set of molecular formulas that are detected together before and after chlorination in each system.
[0170] Based on this, the set of intersection formulas of the three are compared and extracted.
[0171] S6. For the set of intersection molecular formulas obtained in step S5, extract the mass spectrometry characteristic peaks corresponding to each intersection molecular formula that satisfy the aforementioned mass error and signal-to-noise ratio conditions.
[0172] When the same intersection molecular formula corresponds to multiple mass spectrometry characteristic peak records, one mass spectrometry characteristic peak record is retained according to the principle of maximizing the mass spectrometry response intensity within the three systems of EOM, IOM, and AOM, as well as the six sets of data before and after chlorination. The mass spectrometry characteristic peak retained after deduplication is used as the node of the precise mass difference (PMD) reaction transformation network, hereinafter referred to as the PMD network node. For any two PMD network nodes, the measured precise mass difference between the nodes is defined as the measured precise mass difference between the nodes, which is the measured precise mass of the node with the higher mass-to-charge ratio minus the measured precise mass of the node with the lower mass-to-charge ratio.
[0173] The measured precise mass difference was matched with the absolute value of the theoretical precise mass difference corresponding to each preset PMD conversion type. The mass difference matching tolerance was 1.0 ppm. The matching types included the mass difference corresponding to chlorination reaction (+Cl-H), the mass difference corresponding to +Cl-N type conversion (+Cl-N), the mass difference corresponding to chlorination hydroxylation reaction (+Cl+OH), the mass difference corresponding to oxidation reaction (+O, +2O, +3O), the mass difference corresponding to decarboxylation reaction (-CO2), the mass difference corresponding to deoxygenation reaction (-O, -O2), and the mass difference corresponding to sulfur-containing reaction (-SO, -SH2).
[0174] Here, "+" and "-" indicate the direction of change in the element composition of the corresponding PMD conversion type, and are not used as the positive or negative sign of the aforementioned measured precise quality difference.
[0175] The above PMD conversion types are matched based on the measured precise mass difference between mass spectrometry characteristic peaks, and the presence or absence of Cl element in the molecular formula corresponding to the PMD network node is not used as a criterion for determining the chlorination mass difference type.
[0176] TCBQ, 2,6-DBBQ and 2,6-DIBQ were independently quantified by UPLC-MS / MS, and their net concentrations were included in the system-level correlation analysis.
[0177] Using the conversion relationship corresponding to the mass difference as the connection edge, the mass spectrometry characteristic peak node that participates in at least one preset PMD mass difference conversion relationship is defined as the effective node of the PMD network. Based on the mass spectrometry characteristic peak node, the frequency of various mass difference conversion matching and the proportion of chlorinated mass difference conversion nodes are statistically analyzed.
[0178] Among them, the proportion of chlorinated poor mass conversion nodes is calculated as the percentage of the number of effective mass spectrometry characteristic peak nodes participating in +Cl-H, +Cl-N or +Cl+OH poor mass conversions to the total number of effective nodes in the PMD network, and the matching frequency of decarboxylated poor mass conversions is counted according to the number of connection relationships matching -CO2 poor mass conversions.
[0179] In this embodiment, the proportion of chlorinated poor quality conversion nodes in the PMD network is only counted according to the poor quality types of +Cl-H, +Cl-N, and +Cl+OH.
[0180] The amounts of TCBQ, 2,6-DBBQ and 2,6-DIBQ were determined independently by UPLC-MS / MS and used for system-level correlation analysis.
[0181] The proportion of poor-quality chlorinated transformation nodes and the matching frequency of poor-quality decarboxylation transformation nodes are used as the overall criteria for judging the PMD network.
[0182] When the proportion of chlorinated poor quality conversion nodes in the PMD network is ≥40% and the matching frequency of decarboxylated poor quality conversion is ≥3, all effective mass spectrometry characteristic peak nodes in the PMD network are uniformly included in the four-dimensional Boolean logic screening.
[0183] During the screening process, the molecular formula corresponding to each mass spectrometry characteristic peak node and its corresponding molecular parameter record value in the unified data processing flow are read. The four-dimensional Boolean logic condition is:
[0184] The improved aromaticity index is ≥0.5, the oxygen-to-carbon ratio is 0.2 to 0.6, the hydrogen-to-carbon ratio is 0.7 to 1.5, and the number of carbon atoms is 6 to 20. Output a list of candidate precursor molecular formulas.
[0185] The classification of candidate precursor molecular formulas is determined according to the VanKrevelen region division rules mentioned above.
[0186] Example 2:
[0187] This embodiment provides a method for identifying halogenated benzoquinones (HBQs) precursors during the chlorination disinfection process of algal organic matter based on high-resolution mass spectrometry, including the following steps:
[0188] S1. Take the required volumes of EOM high DOC mother liquor, IOM high DOC mother liquor and AOM high DOC mother liquor corresponding to Preparation Example 2 according to the target DOC concentration to establish chlorination reaction systems. Take the corresponding high DOC mother liquor of the same batch as control samples before chlorination. Use ultrapure water to make up to 100 mL to make the DOC 3.0 mg / L.
[0189] Sodium bromide and sodium iodide were added to each chlorination reaction system and the corresponding pre-chlorination control sample, respectively, to allow Br to chlorinate. - Concentration of 300 μg / L, I - The concentration was 300 μg / L, and the initial pH was adjusted to 6.0 using sodium hydroxide or hydrochloric acid;
[0190] Sodium hypochlorite was added to each chlorination reaction system to make the effective chlorine (calculated as Cl2) concentration 15 mg / L. No sodium hypochlorite was added to the control sample before chlorination.
[0191] Each chlorination reaction system was kept at a constant temperature of 25°C for 24 hours under light-protected conditions. After the reaction was completed, the remaining oxidizing halogen equivalent (calculated as Cl2) of the system was determined by N,N-diethyl-p-phenylenediamine spectrophotometry. The remaining oxidizing halogen equivalent was calculated based on the measurement results, and sodium thiosulfate solution was added at 1.5 times the theoretical amount of sodium thiosulfate required for complete reaction to quench the reaction.
[0192] S2. The control sample before chlorination and the three reaction solutions after quenching were pre-filtered through a 0.7μm glass fiber membrane and a 0.45μm glass fiber membrane, respectively, and then filtered through a 0.45μm cellulose acetate membrane. The pH of the samples was adjusted to 2 using hydrochloric acid.
[0193] Subsequently, desalting and enrichment were performed using a styrene-divinylbenzene polymer solid-phase extraction column. Before use, the solid-phase extraction column was activated with methanol and equilibrated with hydrochloric acid solution at pH=2. After the sample was loaded onto the column, it was eluted with 3 column volumes of hydrochloric acid solution at pH=2 to remove inorganic salts. After drying with nitrogen, it was eluted with 20 mL of chromatographic grade methanol. The eluent was concentrated by nitrogen blowing and then reconstituted with a methanol-water mixture at a volume ratio of 50:50 and brought to a final volume of 1.0 mL to obtain the samples to be tested.
[0194] S3. Ultra-high performance liquid chromatography-tandem mass spectrometry (UPLC-MS / MS) in negative ion mode using multiple reaction monitoring (MRM) was employed for targeted quantitative analysis of TCBQ, 2,6-DBBQ, and 2,6-DIBQ in the control sample before chlorination and the sample after chlorination. The concentrations of the samples before and after chlorination were determined, and the net generated concentration was calculated (net generated concentration = concentration after chlorination - concentration before chlorination). Specifically, a Waters ACQUITY UPLC-BEHC18 column (2.1 × 100 mm, 1.7 μm) was used for separation. Mobile phase A was 0.1% formic acid aqueous solution, and mobile phase B was acetonitrile. The gradient elution program was as follows:
[0195] 0–2.0 min, 10% B;
[0196] From 2.0 to 8.0 minutes, the proportion of phase B increased uniformly from 10% to 90%.
[0197] 8.0–10.0 min, 90% B;
[0198] Between 10.0 and 10.1 minutes, the proportion of phase B decreased from 90% to 10%.
[0199] 10.1–13.0 min, 10% B.
[0200] The flow rate was 0.3 mL / min, the column temperature was 40 °C, and the injection volume was 5.0 μL.
[0201] Among them, the MRM quantitative ion pairs (m / z, parent ion / daughter ion) of TCBQ, 2,6-DBBQ and 2,6-DIBQ are 245.8 / 210.8, 264.8 / 236.8 and 358.8 / 330.8, respectively, with corresponding collision energies of 15 eV, 20 eV and 20 eV, respectively;
[0202] TCBQ, 2,6-DBBQ, and 2,6-DIBQ were quantified using the internal standard method. 13 C6-2,6-dichloro-1,4-benzoquinone was used as the internal standard, and the final concentration of the internal standard in the standard curve series and the test samples was 10.0 μg / L.
[0203] The MRM ion pair of the internal standard is 181.0 / 145.0 m / z, corresponding to a collision energy of 15 eV;
[0204] The corresponding standard curve concentration ranges are 0.1–50.0 μg / L, 0.1–50.0 μg / L, and 0.1–50.0 μg / L, respectively. The sample concentration is calculated based on the corresponding standard curve.
[0205] Simultaneously, Fourier transform ion cyclotron resonance mass spectrometry (FT-ICRMS) was used in negative ion electrospray ionization mode to perform non-targeted high-resolution measurements on the control sample before chlorination and the sample after chlorination. The mass spectrometer parameters were set as follows:
[0206] The continuous injection flow rate was 120 μL / h, the capillary inlet voltage was 4.0 kV, the ion accumulation time was 0.06 s, the m / z acquisition range was 100–1600, and the number of time-domain signal superpositions was 300.
[0207] Mass calibration was performed using 10 mmol / L sodium formate, and SRNOM was used as a quality control reference sample for verification to ensure that the mass error was less than 1 ppm, thus obtaining the raw high-resolution mass spectrometry data for each group.
[0208] S4. Following the unified data processing rules described above, molecular formula identification and molecular parameter calculation were performed on the raw high-resolution mass spectrometry data of each group to obtain the DBE, AI_mod, O / C, H / C, number of carbon atoms, NOSC and GFE of each molecular formula, as well as H / C_w, O / C_w, DBE_w, GFE_w, and AI_mod_w. The weighted indices of the control sample before chlorination and the corresponding net concentrations of TCBQ, 2,6-DBBQ and 2,6-DIBQ were included in the unified Spearman rank correlation analysis described above.
[0209] Based on the ρ and p values obtained from the unified Spearman rank correlation analysis above, record the correlation direction and significance between H / C_w, O / C_w, DBE_w, GFE_w, and AI_mod_w and the net generation concentrations of the three HBQs.
[0210] S5. First, compare the mass spectrometry results of the control samples before chlorination and the samples after chlorination of the EOM, IOM and AOM reaction systems corresponding to Preparation Example 2, and extract the set of molecular formulas that are detected together before and after chlorination in each system.
[0211] Based on this, the set of intersection molecular formulas that coexist among the three is compared and extracted; S6, for the extracted set of intersection molecular formulas, a PMD network is constructed according to the PMD transformation type and network determination rules in step S6 of Example 1.
[0212] Finally, the molecular formulas and molecular parameter records corresponding to the mass spectrometry characteristic peak nodes in the PMD network are read, and four-dimensional Boolean logic is applied for filtering. The four-dimensional Boolean logic conditions are:
[0213] The improved aromaticity index is ≥0.5, the oxygen-to-carbon ratio is 0.2 to 0.6, the hydrogen-to-carbon ratio is 0.7 to 1.5, and the number of carbon atoms is 6 to 20. Output a list of candidate precursor molecular formulas.
[0214] Example 3:
[0215] This embodiment provides a method for identifying halogenated benzoquinones (HBQs) precursors during the chlorination disinfection process of algal organic matter based on high-resolution mass spectrometry, including the following steps:
[0216] S1. Take the required volumes of EOM high DOC mother liquor, IOM high DOC mother liquor and AOM high DOC mother liquor corresponding to Preparation Example 3 according to the target DOC concentration to establish chlorination reaction systems. Take the corresponding high DOC mother liquor from the same batch as control samples before chlorination. Use ultrapure water to make up to 100 mL to make the DOC 1.0 mg / L.
[0217] Sodium bromide and sodium iodide were added to each chlorination reaction system and the corresponding pre-chlorination control sample, respectively, to allow Br to chlorinate. - Concentration of 100 μg / L, I - The concentration was 100 μg / L, and the initial pH was adjusted to 3.0 using sodium hydroxide or hydrochloric acid;
[0218] Sodium hypochlorite was added to each chlorination reaction system to make the effective chlorine (calculated as Cl2) concentration 5.0 mg / L. No sodium hypochlorite was added to the control sample before chlorination.
[0219] Each chlorination reaction system was kept at a constant temperature of 25°C for 2 hours under light-protected conditions. After the reaction was completed, the remaining oxidizing halogen equivalent (calculated as Cl2) of the system was determined by N,N-diethyl-p-phenylenediamine spectrophotometry. The remaining oxidizing halogen equivalent was calculated based on the measurement results, and sodium thiosulfate solution was added at 1.5 times the theoretical amount of sodium thiosulfate required for complete reaction to quench the reaction.
[0220] S2. The control sample before chlorination and the three reaction solutions after quenching were pre-filtered through a 0.7μm glass fiber membrane and a 0.45μm glass fiber membrane, respectively, and then filtered through a 0.45μm cellulose acetate membrane. The pH of the samples was adjusted to 2 using hydrochloric acid.
[0221] Subsequently, desalting and enrichment were performed using a styrene-divinylbenzene polymer solid-phase extraction column. Before use, the solid-phase extraction column was activated with methanol and equilibrated with hydrochloric acid solution at pH=2. After the sample was loaded onto the column, it was eluted with 3 column volumes of hydrochloric acid solution at pH=2 to remove inorganic salts. After drying with nitrogen, it was eluted with 20 mL of chromatographic grade methanol. The eluent was concentrated by nitrogen blowing and then reconstituted with a methanol-water mixture at a volume ratio of 50:50 and brought to a final volume of 1.0 mL to obtain the samples to be tested.
[0222] S3. Ultra-high performance liquid chromatography-tandem mass spectrometry (UPLC-MS / MS) in negative ion mode using multiple reaction monitoring (MRM) was employed for targeted quantitative analysis of TCBQ, 2,6-DBBQ, and 2,6-DIBQ in the control sample before chlorination and the sample after chlorination. The concentrations of the samples before and after chlorination were determined, and the net generated concentration was calculated (net generated concentration = concentration after chlorination - concentration before chlorination). Specifically, a Waters ACQUITY UPLC-BEHC18 column (2.1 × 100 mm, 1.7 μm) was used for separation. Mobile phase A was 0.1% formic acid aqueous solution, and mobile phase B was acetonitrile. The gradient elution program was as follows:
[0223] 0–2.0 min, 10% B;
[0224] From 2.0 to 8.0 minutes, the proportion of phase B increased uniformly from 10% to 90%.
[0225] 8.0–10.0 min, 90% B;
[0226] Between 10.0 and 10.1 minutes, the proportion of phase B decreased from 90% to 10%.
[0227] 10.1–13.0 min, 10% B.
[0228] The flow rate was 0.3 mL / min, the column temperature was 40 °C, and the injection volume was 5.0 μL.
[0229] Among them, the MRM quantitative ion pairs (m / z, parent ion / daughter ion) of TCBQ, 2,6-DBBQ and 2,6-DIBQ are 245.8 / 210.8, 264.8 / 236.8 and 358.8 / 330.8, respectively, with corresponding collision energies of 15 eV, 20 eV and 20 eV, respectively;
[0230] TCBQ, 2,6-DBBQ, and 2,6-DIBQ were quantified using the internal standard method. 13 C6-2,6-dichloro-1,4-benzoquinone was used as the internal standard, and the final concentration of the internal standard in the standard curve series and the test samples was 10.0 μg / L.
[0231] The MRM ion pair of the internal standard is 181.0 / 145.0 m / z, corresponding to a collision energy of 15 eV;
[0232] The corresponding standard curve concentration ranges are 0.1–50.0 μg / L, 0.1–50.0 μg / L, and 0.1–50.0 μg / L, respectively. The sample concentration is calculated based on the corresponding standard curve.
[0233] Simultaneously, Fourier transform ion cyclotron resonance mass spectrometry (FT-ICRMS) was used in negative ion electrospray ionization mode to perform non-targeted high-resolution measurements on the control sample before chlorination and the sample after chlorination. The mass spectrometer parameters were set as follows:
[0234] The continuous injection flow rate was 100 μL / h, the capillary inlet voltage was 3.5 kV, the ion accumulation time was 0.05 s, the m / z acquisition range was 100–1600, and the number of time-domain signal superpositions was 200.
[0235] Mass calibration was performed using 10 mmol / L sodium formate, and SRNOM was used as a quality control reference sample for verification to ensure that the mass error was less than 1 ppm, thus obtaining the raw high-resolution mass spectrometry data for each group.
[0236] S4. According to the above unified data processing rules, molecular formula identification and molecular parameter calculation are performed on each group of high-resolution mass spectrometry raw data to obtain DBE, AI_mod, O / C, H / C, number of carbon atoms, NOSC and GFE, as well as H / C_w, O / C_w, DBE_w, GFE_w, and AI_mod_w for each molecular formula.
[0237] The weighted indices and corresponding net concentrations of TCBQ, 2,6-DBBQ and 2,6-DIBQ of the pre-chlorination control sample were included in the above unified Spearman rank correlation analysis.
[0238] Based on the ρ and p values obtained from the unified Spearman rank correlation analysis above, record the correlation direction and significance between H / C_w, O / C_w, DBE_w, GFE_w, and AI_mod_w and the net generation concentrations of the three HBQs.
[0239] S5. First, compare the mass spectrometry results of the control samples before chlorination and the samples after chlorination of the EOM, IOM and AOM reaction systems corresponding to Preparation Example 3, and extract the set of molecular formulas that are detected together before and after chlorination in each system.
[0240] Based on this, the set of intersection formulas of the three are compared and extracted.
[0241] S6. For the extracted set of intersection molecular formulas, construct a PMD network according to the PMD transformation type and network determination rules in step S6 of Example 1;
[0242] Finally, the molecular formulas and molecular parameter records corresponding to the mass spectrometry characteristic peak nodes in the PMD network are read, and four-dimensional Boolean logic is applied for filtering. The four-dimensional Boolean logic conditions are:
[0243] The improved aromaticity index is ≥0.5, the oxygen-to-carbon ratio is 0.2 to 0.6, the hydrogen-to-carbon ratio is 0.7 to 1.5, and the number of carbon atoms is 6 to 20. Output a list of candidate precursor molecular formulas.
[0244] Example 4:
[0245] This embodiment provides a method for identifying halogenated benzoquinones (HBQs) precursors during the chlorination disinfection process of algal organic matter based on high-resolution mass spectrometry, including the following steps:
[0246] S1. Take the EOM high DOC mother liquor, IOM high DOC mother liquor and AOM high DOC mother liquor from the same batch retained during the parallel scale-up preparation of Example 4, and establish parallel chlorination reaction systems independently. Take the corresponding EOM high DOC mother liquor, IOM high DOC mother liquor and AOM high DOC mother liquor from the same batch as control samples before chlorination.
[0247] Ultrapure water was used to bring the volume to 100 mL, ensuring that the DOC concentration in both the reaction system and the control sample was 4.0 mg / L. Sodium bromide and sodium iodide were added to each chlorination reaction system and the corresponding pre-chlorination control sample, respectively, to ensure that the Br... - Concentration of 200 μg / L, I - The concentration was 400 μg / L, and the initial pH was adjusted to 7.0 using sodium hydroxide or hydrochloric acid;
[0248] Sodium hypochlorite was added to each chlorination reaction system to make the effective chlorine (calculated as Cl2) concentration 10 mg / L. No sodium hypochlorite was added to the control sample before chlorination.
[0249] Each chlorination reaction system was kept at a constant temperature of 25°C for 36 hours under light-protected conditions. After the reaction was completed, the remaining oxidizing halogen equivalent (calculated as Cl2) of the system was determined by N,N-diethyl-p-phenylenediamine spectrophotometry. The remaining oxidizing halogen equivalent was calculated based on the measurement results, and sodium thiosulfate solution was added at 1.5 times the theoretical amount of sodium thiosulfate required for complete reaction to quench the reaction.
[0250] S2. The control sample before chlorination and the three reaction solutions after quenching were pre-filtered through a 0.7μm glass fiber membrane and a 0.45μm glass fiber membrane, respectively, and then filtered through a 0.45μm cellulose acetate membrane. The pH of the samples was adjusted to 2 using hydrochloric acid.
[0251] Subsequently, desalting and enrichment were performed using a styrene-divinylbenzene polymer solid-phase extraction column. Before use, the solid-phase extraction column was activated with methanol and equilibrated with hydrochloric acid solution at pH=2. After the sample was loaded onto the column, it was eluted with 3 column volumes of hydrochloric acid solution at pH=2 to remove inorganic salts. After drying with nitrogen, it was eluted with 20 mL of chromatographic grade methanol. The eluent was concentrated by nitrogen blowing and then reconstituted with a methanol-water mixture at a volume ratio of 50:50 and brought to a final volume of 1.0 mL to obtain the samples to be tested.
[0252] S3. Ultra-high performance liquid chromatography-tandem mass spectrometry (UPLC-MS / MS) in negative ion mode using multiple reaction monitoring (MRM) was employed for targeted quantitative analysis of TCBQ, 2,6-DBBQ, and 2,6-DIBQ in the control sample before chlorination and the sample after chlorination. The concentrations of the samples before and after chlorination were determined, and the net generated concentration was calculated (net generated concentration = concentration after chlorination - concentration before chlorination). Specifically, a Waters ACQUITY UPLC-BEHC18 column (2.1 × 100 mm, 1.7 μm) was used for separation. Mobile phase A was 0.1% formic acid aqueous solution, and mobile phase B was acetonitrile. The gradient elution program was as follows:
[0253] 0–2.0 min, 10% B;
[0254] From 2.0 to 8.0 minutes, the proportion of phase B increased uniformly from 10% to 90%.
[0255] 8.0–10.0 min, 90% B;
[0256] Between 10.0 and 10.1 minutes, the proportion of phase B decreased from 90% to 10%.
[0257] 10.1–13.0 min, 10% B.
[0258] The flow rate was 0.3 mL / min, the column temperature was 40 °C, and the injection volume was 5.0 μL.
[0259] Among them, the MRM quantitative ion pairs (m / z, parent ion / daughter ion) of TCBQ, 2,6-DBBQ and 2,6-DIBQ are 245.8 / 210.8, 264.8 / 236.8 and 358.8 / 330.8, respectively, with corresponding collision energies of 15 eV, 20 eV and 20 eV, respectively;
[0260] TCBQ, 2,6-DBBQ, and 2,6-DIBQ were quantified using the internal standard method. 13 C6-2,6-dichloro-1,4-benzoquinone was used as the internal standard, and the final concentration of the internal standard in the standard curve series and the test samples was 10.0 μg / L.
[0261] The MRM ion pair of the internal standard is 181.0 / 145.0 m / z, corresponding to a collision energy of 15 eV;
[0262] The corresponding standard curve concentration ranges are 0.1–50.0 μg / L, 0.1–50.0 μg / L, and 0.1–50.0 μg / L, respectively. The sample concentration is calculated based on the corresponding standard curve.
[0263] Simultaneously, Fourier transform ion cyclotron resonance mass spectrometry (FT-ICRMS) was used in negative ion electrospray ionization mode to perform non-targeted high-resolution measurements on the control sample before chlorination and the sample after chlorination. The mass spectrometer parameters were set as follows:
[0264] The continuous injection flow rate was 130 μL / h, the capillary inlet voltage was 4.2 kV, the ion accumulation time was 0.08 s, the m / z acquisition range was 100–1600, and the number of time-domain signal superpositions was 350.
[0265] Mass calibration was performed using 10 mmol / L sodium formate, and SRNOM was used as a quality control reference sample for verification to ensure that the mass error was less than 1 ppm, thus obtaining the raw high-resolution mass spectrometry data for each group.
[0266] S4. According to the above unified data processing rules, molecular formula identification and molecular parameter calculation are performed on each group of high-resolution mass spectrometry raw data to obtain DBE, AI_mod, O / C, H / C, number of carbon atoms, NOSC and GFE, as well as H / C_w, O / C_w, DBE_w, GFE_w, and AI_mod_w for each molecular formula.
[0267] The weighted indices and corresponding net concentrations of TCBQ, 2,6-DBBQ and 2,6-DIBQ of the pre-chlorination control sample were included in the above unified Spearman rank correlation analysis.
[0268] Based on the ρ and p values obtained from the unified Spearman rank correlation analysis above, record the correlation direction and significance between H / C_w, O / C_w, DBE_w, GFE_w, and AI_mod_w and the net generation concentrations of the three HBQs.
[0269] S5. First, compare the mass spectrometry results of the control samples before chlorination and the samples after chlorination of the EOM, IOM and AOM reaction systems corresponding to Preparation Example 4, and extract the set of molecular formulas that are detected together before and after chlorination in each system.
[0270] Based on this, the set of intersection formulas of the three are compared and extracted.
[0271] S6. For the extracted set of intersection molecular formulas, construct a PMD network according to the PMD transformation type and network determination rules in step S6 of Example 1;
[0272] Finally, the molecular formulas and molecular parameter records corresponding to the mass spectrometry characteristic peak nodes in the PMD network are read, and four-dimensional Boolean logic is applied for filtering. The four-dimensional Boolean logic conditions are:
[0273] The improved aromaticity index is ≥0.5, the oxygen-to-carbon ratio is 0.2 to 0.6, the hydrogen-to-carbon ratio is 0.7 to 1.5, and the number of carbon atoms is 6 to 20. Output a list of candidate precursor molecular formulas.
[0274] Example 5:
[0275] This embodiment provides a method for identifying halogenated benzoquinones (HBQs) precursors during the chlorination disinfection process of algal organic matter based on high-resolution mass spectrometry, including the following steps:
[0276] S1. Take the required volumes of EOM high DOC mother liquor, IOM high DOC mother liquor and AOM high DOC mother liquor corresponding to Preparation Example 5 according to the target DOC concentration to establish chlorination reaction systems. Take the corresponding high DOC mother liquor from the same batch as control samples before chlorination. Use ultrapure water to make up to 100 mL to make the DOC 2.0 mg / L.
[0277] Sodium bromide and sodium iodide were added to each chlorination reaction system and the corresponding pre-chlorination control sample, respectively, to allow Br to chlorinate. - Concentration of 400 μg / L, I - The concentration was 200 μg / L, and the initial pH was adjusted to 5.0 using sodium hydroxide or hydrochloric acid;
[0278] Sodium hypochlorite was added to each chlorination reaction system to make the effective chlorine (calculated as Cl2) concentration 20 mg / L. No sodium hypochlorite was added to the control sample before chlorination.
[0279] Each chlorination reaction system was kept at a constant temperature of 25°C for 12 hours under light-protected conditions. After the reaction was completed, the remaining oxidizing halogen equivalent (calculated as Cl2) of the system was determined by N,N-diethyl-p-phenylenediamine spectrophotometry. The remaining oxidizing halogen equivalent was calculated based on the measurement results, and sodium thiosulfate solution was added at 1.5 times the theoretical amount of sodium thiosulfate required for complete reaction to quench the reaction.
[0280] S2. The control sample before chlorination and the three reaction solutions after quenching were pre-filtered through a 0.7μm glass fiber membrane and a 0.45μm glass fiber membrane, respectively, and then filtered through a 0.45μm cellulose acetate membrane. The pH of the samples was adjusted to 2 using hydrochloric acid.
[0281] Subsequently, desalting and enrichment were performed using a styrene-divinylbenzene polymer solid-phase extraction column. Before use, the solid-phase extraction column was activated with methanol and equilibrated with hydrochloric acid solution at pH=2. After the sample was loaded onto the column, it was eluted with 3 column volumes of hydrochloric acid solution at pH=2 to remove inorganic salts. After drying with nitrogen, it was eluted with 20 mL of chromatographic grade methanol. The eluent was concentrated by nitrogen blowing and then reconstituted with a methanol-water mixture at a volume ratio of 50:50 and brought to a final volume of 1.0 mL to obtain the samples to be tested.
[0282] S3. Ultra-high performance liquid chromatography-tandem mass spectrometry (UPLC-MS / MS) in negative ion mode using multiple reaction monitoring (MRM) was employed for targeted quantitative analysis of TCBQ, 2,6-DBBQ, and 2,6-DIBQ in the control sample before chlorination and the sample after chlorination. The concentrations of the samples before and after chlorination were determined, and the net generated concentration was calculated (net generated concentration = concentration after chlorination - concentration before chlorination). Specifically, a Waters ACQUITY UPLC-BEHC18 column (2.1 × 100 mm, 1.7 μm) was used for separation. Mobile phase A was 0.1% formic acid aqueous solution, and mobile phase B was acetonitrile. The gradient elution program was as follows:
[0283] 0–2.0 min, 10% B;
[0284] From 2.0 to 8.0 minutes, the proportion of phase B increased uniformly from 10% to 90%.
[0285] 8.0–10.0 min, 90% B;
[0286] Between 10.0 and 10.1 minutes, the proportion of phase B decreased from 90% to 10%.
[0287] 10.1–13.0 min, 10% B.
[0288] The flow rate was 0.3 mL / min, the column temperature was 40 °C, and the injection volume was 5.0 μL.
[0289] Among them, the MRM quantitative ion pairs (m / z, parent ion / daughter ion) of TCBQ, 2,6-DBBQ and 2,6-DIBQ are 245.8 / 210.8, 264.8 / 236.8 and 358.8 / 330.8, respectively, with corresponding collision energies of 15 eV, 20 eV and 20 eV, respectively;
[0290] TCBQ, 2,6-DBBQ, and 2,6-DIBQ were quantified using the internal standard method. 13 C6-2,6-dichloro-1,4-benzoquinone was used as the internal standard, and the final concentration of the internal standard in the standard curve series and the test samples was 10.0 μg / L.
[0291] The MRM ion pair of the internal standard is 181.0 / 145.0 m / z, corresponding to a collision energy of 15 eV;
[0292] The corresponding standard curve concentration ranges are 0.1–50.0 μg / L, 0.1–50.0 μg / L, and 0.1–50.0 μg / L, respectively. The sample concentration is calculated based on the corresponding standard curve.
[0293] Simultaneously, Fourier transform ion cyclotron resonance mass spectrometry (FT-ICRMS) was used in negative ion electrospray ionization mode to perform non-targeted high-resolution measurements on the control sample before chlorination and the sample after chlorination. The mass spectrometer parameters were set as follows:
[0294] The continuous injection flow rate was 110 μL / h, the capillary inlet voltage was 3.8 kV, the ion accumulation time was 0.06 s, the m / z acquisition range was 100–1600, and the number of time-domain signal superpositions was 250.
[0295] Mass calibration was performed using 10 mmol / L sodium formate, and SRNOM was used as a quality control reference sample for verification to ensure that the mass error was less than 1 ppm, thus obtaining the raw high-resolution mass spectrometry data for each group.
[0296] S4. According to the above unified data processing rules, molecular formula identification and molecular parameter calculation are performed on each group of high-resolution mass spectrometry raw data to obtain DBE, AI_mod, O / C, H / C, number of carbon atoms, NOSC and GFE, as well as H / C_w, O / C_w, DBE_w, GFE_w, and AI_mod_w for each molecular formula.
[0297] The weighted indices and corresponding net concentrations of TCBQ, 2,6-DBBQ and 2,6-DIBQ of the pre-chlorination control sample were included in the above unified Spearman rank correlation analysis.
[0298] Based on the ρ and p values obtained from the unified Spearman rank correlation analysis above, record the correlation direction and significance between H / C_w, O / C_w, DBE_w, GFE_w, and AI_mod_w and the net generation concentrations of the three HBQs.
[0299] S5. First, compare the mass spectrometry results of the control samples before chlorination and the samples after chlorination of the EOM, IOM and AOM reaction systems corresponding to Preparation Example 5, and extract the set of molecular formulas that are detected together before and after chlorination in each system.
[0300] Based on this, the set of intersection formulas of the three are compared and extracted;
[0301] S6. For the extracted set of intersection molecular formulas, construct a PMD network according to the PMD transformation type and network determination rules in step S6 of Example 1;
[0302] Finally, the molecular formulas and molecular parameter records corresponding to the mass spectrometry characteristic peak nodes in the PMD network are read, and four-dimensional Boolean logic is applied for filtering. The four-dimensional Boolean logic conditions are:
[0303] The improved aromaticity index is ≥0.5, the oxygen-to-carbon ratio is 0.2 to 0.6, the hydrogen-to-carbon ratio is 0.7 to 1.5, and the number of carbon atoms is 6 to 20. Output a list of candidate precursor molecular formulas.
[0304] Comparative Example 1:
[0305] Compared with Example 2, the difference is that in the data processing logic of steps S4 and S6, Spearman's rank correlation analysis was not performed, the PMD network was not constructed, and four-dimensional Boolean logic constraints were not implemented. The molecular formula was only classified and statistically analyzed according to the Van Krevelen region division rule. The remaining steps and parameters are the same as in Example 2.
[0306] Comparative Example 2:
[0307] Compared with Example 2, the difference is that in the data processing logic of step S6, the limitation of "improved aromaticity index ≥ 0.5" is removed, and the range of hydrogen-carbon atom ratio is adjusted to 0.1 to 2.5; the remaining steps and parameters are the same as in Example 2.
[0308] Comparative Example 3:
[0309] Compared with Example 2, the difference is that in step S3, the Fourier transform ion cyclotron resonance mass spectrometer is replaced with a high performance liquid chromatography-quadrupole time-of-flight mass spectrometer for measurement. The mass spectrometry acquisition parameters and mass calibration parameters adopt the electrospray negative ion source mode, the capillary voltage is 2.5kV, the m / z acquisition range is 100 to 1000, and the mass calibration uses sodium formate cluster ions for external standard calibration.
[0310] When performing molecular formula identification on high performance liquid chromatography-quadrupole time-of-flight mass spectrometry data, the molecular formula identification quality error limit was set to 10.0 ppm; the remaining steps and parameters were the same as in Example 2.
[0311] Comparative Example 4:
[0312] Compared with Example 2, the difference is that in step S1, only the single component of mixed AOM-2 was extracted to establish a single chlorination reaction system, and no parallel control group of EOM-2 and IOM-2 was set up.
[0313] Furthermore, in step S5, the extraction step of the intersection of the three molecular formulas EOM, IOM and AOM is not performed. Instead, the single-component data of the mixed AOM-2 is directly introduced into the subsequent network construction and algorithm operation. The remaining steps and parameters are the same as in Example 2.
[0314] Test Example 1:
[0315] 1. The candidate precursor molecular formula lists output by EOM, IOM and AOM intersection molecular formula extraction, PMD network construction and four-dimensional Boolean logic screening of Examples 1 to 5 were collected as test objects; the unified Spearman-level correlation analysis of Examples 1 to 5 was carried out independently according to the aforementioned system-level correlation analysis rules, and this analysis was not used as a screening condition for the candidate precursor molecular formula list.
[0316] 2. The molecular formula data in each embodiment list are statistically analyzed using data processing software, and the total number of candidate precursor molecular formulas detected in the corresponding embodiments is calculated and recorded.
[0317] 3. Based on the identified molecular formula, calculate the average molecular mass, average double bond equivalent, and average improved aromaticity index of the candidate precursors in the list.
[0318] 4. Summarize and compare the above statistical results, and observe the changes in the number of candidate precursor molecular formulas, average molecular mass, average double bond equivalent, and average improved aromaticity index under different physicochemical conditions and mass spectrometry parameters.
[0319] Table 1. Parameter statistics of the candidate precursor lists output in Examples 1 to 5.
[0320] Group Number of candidate precursor molecular formulas (number of) Average molecular weight (Da) Average double bond equivalent Average Improved Aromaticity Index Example 1 28 345.12 9.8 0.62 Example 2 23 332.45 10.1 0.58 Example 3 31 358.91 9.4 0.55 Example 4 25 327.68 10.5 0.61 Example 5 26 341.05 9.7 0.59
[0321] Based on the statistical data in Table 1 and Figure 1 The scatter distribution of the samples shows that the parameters of the candidate precursors output in Examples 1 to 5 differ under different reaction and instrument conditions; from Figure 1 (a) It can be seen that the number of candidate precursor molecular formulas output by each embodiment ranges from 23 to 31, with average molecular weights ranging from 327.68 to 358.91 Da. Furthermore, the number of molecular formulas shows a positive trend with increasing average molecular weight. Figure 1 (b) It can be seen that the average double bond equivalent of the corresponding molecules in each embodiment is concentrated between 9.4 and 10.5, and the average improved aromaticity index fluctuates between 0.55 and 0.62. The above data show that changes in the combination of physicochemical conditions and mass spectrometry parameters will cause fluctuations in the number of candidate precursors and molecular parameters, but the overall distribution is still within the characteristic range of high unsaturation and high aromaticity.
[0322] The above test results show that the precursor identification method provided by the present invention has stability and feasibility. Under the unified process of intersection molecular formula extraction, PMD network construction and four-dimensional Boolean logic screening, even if the initial processing conditions of different embodiments are different, the method can still continuously output a list of molecular formulas that meet the characteristics of HBQs precursors. The screened candidate molecules all meet the structural requirements of high double bond equivalent and high aromaticity (AI_mod≥0.5), indicating that the multi-dimensional screening conditions and numerical boundaries set by the present invention can effectively eliminate interference data in the background of complex organic matter and achieve objective identification of target precursor molecules.
[0323] Test Example 2:
[0324] 1. Using the high-DOC mother liquor corresponding to Preparation Example 1, separate chlorination reaction systems with a total volume of 500 mL were prepared. The DOC concentration of each chlorination reaction system was 3.0 mg / L, the available chlorine (as Cl2) concentration was 15.0 mg / L, and the Br content was [missing information]. - Concentration of 200 μg / L, I - The concentration was 50 μg / L, the initial pH was 7.0, and the reaction time points were set at 2h, 6h, 12h, 24h and 48h under constant temperature and dark conditions at 25℃.
[0325] Separately, a 500 mL pre-chlorination control sample was prepared from the same batch of EOM high-DOC stock solution. The DOC concentration of the pre-chlorination control sample was 3.0 mg / L, and the Br content was [missing information]. - Concentration of 200 μg / L, I - The concentration was 50 μg / L, the initial pH was 7.0, and no sodium hypochlorite was added;
[0326] Independent parallel reaction systems were used at each time point. TCBQ, 2,6-DBBQ and 2,6-DIBQ were determined according to the same quenching and UPLC-MS / MS quantitative procedure, and their net concentrations were calculated using the same method.
[0327] 2. Extract the corresponding mass spectrometry characteristic peaks from the set of molecular formulas detected together by the control sample before chlorination and the samples at each time point;
[0328] For multiple mass spectrometry characteristic peak records of the same molecular formula in the control sample before chlorination and the sample at the corresponding time point, one mass spectrometry characteristic peak record is retained according to the principle of the maximum mass spectrometry response intensity in the two sets of data. The mass spectrometry characteristic peak retained after deduplication is used as the PMD network node to construct the corresponding PMD network. The number of effective nodes participating in the mass difference conversion of -CO2 and the number of effective nodes participating in the mass difference conversion of +Cl-H, +Cl-N or +Cl+OH are extracted from the network at each time point, and their distribution ratio in the total number of effective nodes in the PMD network is calculated respectively.
[0329] Among them, decarboxylation poor quality conversion nodes are determined according to their participation in -CO2 poor quality conversion, and chlorination poor quality conversion nodes are determined according to their participation in +Cl-H, +Cl-N, or +Cl+OH poor quality conversion.
[0330] In this test case, the proportion of decarboxylation-related poor quality conversion nodes is calculated as the percentage of the number of effective nodes participating in -CO2 poor quality conversion to the total number of effective nodes in the PMD network. This is a different statistical indicator from the decarboxylation-related poor quality conversion matching frequency counted according to the number of connection relationships in step S6 of Example 1.
[0331] 3. Based on the difference in node proportion between adjacent time points and the time span, calculate the rate of change of node proportion in each interval. In the interval from 0 to 2h, the proportion of decarboxylation poor quality conversion nodes and the proportion of chlorination poor quality conversion nodes in 0h are defined as 0, and the rate of change of node proportion is calculated based on this defined value.
[0332] 4. Summarize the node proportions and the rate of change of node proportions, and record the proportions of decarboxylation poor quality conversion nodes and chlorination poor quality conversion nodes at different reaction times, as well as the rate of change of node proportions within each interval.
[0333] Table 2. Statistical table of the proportion of decarboxylation poor conversion nodes and chlorination poor conversion nodes in the reaction system with reaction time and the rate of change of the proportion of nodes.
[0334] Reaction time (h) Data source Percentage of poor-quality decarboxylation conversion nodes (%) Percentage of poor-quality chlorinated conversion nodes (%) Rate of change in the proportion of poor-quality decarboxylation conversion nodes (percentage points / h) Rate of change in the percentage of poor-quality chlorinated conversion nodes (percentage points / h) 2 Sampling time point 1 13.5 1.8 6.75 0.90 6 Sampling time point 2 28.4 4.2 3.73 0.60 12 Sampling time point 3 41.2 7.4 2.13 0.53 24 Sampling time point 4 52.8 28.5 0.97 1.76 48 Sampling time point 5 58.1 71.9 0.22 1.81
[0335] Combining the data in Table 2 with Figure 2 The evolution trend shows that the proportion and rate of change of decarboxylation and chlorination quality difference conversion nodes differ at different reaction stages; for example... Figure 2 (a) and Figure 2 As shown in (b), during the 0-12h stage, the proportion of decarboxylation conversion nodes increased from 13.5% at 2h to 41.2% at 12h, with a rate of change between 2.13 and 6.75 percentage points / h; during this stage, the proportion of chlorination conversion nodes increased from 1.8% to 7.4%, with a rate of change between 0.53 and 0.90 percentage points / h; during the 12h-48h stage, the growth rate of the proportion of decarboxylation conversion nodes slowed down, with the rate of change decreasing to 0.97 and 0.22 percentage points / h, respectively; the growth rate of the proportion of chlorination conversion nodes increased, with the rate of change increasing to 1.76 and 1.81 percentage points / h, and the proportion of nodes reaching 71.9% at 48h; the data indicate that as the reaction time increases, the quality difference between decarboxylation and chlorination in the network nodes shows an evolutionary characteristic of decarboxylation being dominant in the early stage and chlorination increasing in the later stage.
[0336] The above test results illustrate the basis for setting the PMD molecular network feature transformation threshold in this invention. The data reflects that the process of organic matter generating HBQs involves decarboxylation and chlorination reactions, and chlorination-type poor quality transformation nodes only occupy a high proportion in the network in the later stage of the reaction. Step S6 of this invention uses the proportion of chlorination-type poor quality transformation nodes ≥40% and the decarboxylation matching frequency as a joint preset threshold, which is consistent with the data evolution law of the above reaction process. This threshold setting can exclude background molecules that did not fully participate in the chlorination reaction in the early stage of the reaction, and extract network nodes that actually participate in the halogenation process in the middle and later stages of the reaction for four-dimensional Boolean logic screening, thereby ensuring the objectivity of the precursor identification results.
[0337] Test Example 3:
[0338] 1. Based on the parameter settings of Example 2, the extraction of extracellular organic matter, intracellular organic matter and mixed organic matter, the construction of chlorination reaction system, high-resolution mass spectrometry determination and data processing were repeatedly performed on the same batch of water samples, and 6 independent replicate experimental groups were set up in parallel.
[0339] 2. Collect the list of candidate precursor molecular formulas output from the above 6 sets of repeated experimental groups after PMD network construction and four-dimensional Boolean logic screening, and use it as the data object for repeatability and precision analysis.
[0340] 3. By integrating the data from the 6 candidate lists, the total number of candidate molecular formulas detected in each replicate group was counted, and the number of molecular formulas detected in all 6 groups of data was extracted.
[0341] 4. Based on the identification results, calculate the average molecular mass, average double bond equivalent, and average improved aromaticity index of all molecular formulas in the candidate list of each repeat group, and summarize and record the fluctuation range of each data and the overlap between groups.
[0342] Table 3. Statistical table of parameters measured repeatedly and independently under the method of Example 2.
[0343] Repeat group Total number of candidate precursor molecular formulas (number) Number of molecular formulas detected in all 6 groups Average molecular weight (Da) Average double bond equivalent Average Improved Aromaticity Index Repeat group 1 24 21 331.45 10.2 0.59 Repeat group 2 26 21 336.88 10.4 0.61 Repeat group 3 22 21 328.74 9.8 0.57 Repeat group 4 25 21 335.12 10.3 0.59 Repeat group 5 23 21 330.15 9.9 0.58 Repeat group 6 24 21 333.60 10.2 0.60
[0344] Combining the data in Table 3 with Figure 3 As a result, the candidate precursor parameters output from the six sets of parallel tests exhibited a concentrated distribution characteristic; from Figure 3 The bar chart in (a) shows that the total number of candidate molecular formulas output by the six replicate experiments ranged from 22 to 26, with 21 molecular formulas being detected in each group, representing a common detection rate between 80.77% and 95.45%. Figure 3(b) The line graph shows that the average molecular weight of each group is distributed around the mean reference line, with specific values ranging from 328.74 to 336.88 Da. In addition, Table 3 shows that the average double bond equivalent of each group is between 9.8 and 10.4, and the average improved aromaticity index is between 0.57 and 0.61. The data indicate that, under the same sample and operating conditions, the molecular formula list obtained by multiple independent determinations has an overlap of more than 80%, and the molecular weight, unsaturation degree and aromaticity index of the corresponding molecules have small differences.
[0345] The above test results demonstrate the data reproducibility and precision of the method of this invention. The raw data from non-targeted high-resolution mass spectrometry suffers from background interference and random fluctuations. This invention forcibly filters the raw data by defining the extraction rules for PMD molecular network feature nodes and using four-dimensional Boolean logic screening boundaries. Data from six parallel tests show that after the above screening steps, the candidate lists output by each group can retain a core molecule set with extremely high overlap, and the multidimensional physicochemical parameters of molecules screened from different batches remain stable. This indicates that the parameter thresholds and processing logic set in this invention can eliminate random errors caused by instrument measurements, ensuring the reliability of precursor identification results.
[0346] Test Example 4:
[0347] 1. Extract the list of candidate precursor molecular formulas from Example 2 and Comparative Example 2, as well as the set of molecular formulas classified and statistically analyzed in Comparative Example 1 according to the Van Krevelen region division rule, as the data objects for this comparative experiment.
[0348] 2. Use data statistics software to classify and summarize the total number of molecular formulas in the above three output lists, and record the total number of detected molecular formulas.
[0349] 3. Based on the molecular formulas identified by high-resolution mass spectrometry, calculate the improved aromaticity index (AI_mod) and the hydrogen-to-carbon ratio (H / C) for all molecular formulas in each list.
[0350] 4. Define molecular formulas with AI_mod < 0.5 as low AI_mod molecular formulas and molecular formulas with H / C > 1.5 as high H / C molecular formulas. Count the number of molecular formulas in each category and the number of retained molecular formulas that satisfy AI_mod ≥ 0.5.
[0351] 5. Compare the total number of molecular formulas detected in each group, the number of AI_mod≥0.5, the number of AI_mod<0.5, and the number of H / C>1.5.
[0352] Table 4. Statistical table of screening results parameters for different data processing methods in Example 2 and the comparative example.
[0353] Group Total number of molecular formulas detected. AI_mod≥0.5 Quantity (pieces) AI_mod<0.5 Quantity (pieces) H / C > 1.5 Quantity (pieces) Example 2 23 23 0 0 Comparative Example 1 3415 876 2539 1724 Comparative Example 2 458 122 336 215
[0354] Combined with Table 4 Figure 4 The number of molecular formulas and the proportion of parameters output by different data processing methods vary; Figure 4 As can be seen from the columnar distribution in (a), the total number of molecular formulas detected in Comparative Example 1, Comparative Example 2 and Example 2 were 3415, 458 and 23, respectively; Figure 4 (b) According to the specific values in Table 4, the proportions of low AI_mod (<0.5) and high H / C (>1.5) molecular formulas in Comparative Example 1 were 74.35% and 50.48%, respectively; the proportions of these two types of molecular formulas in Comparative Example 2 were 73.40% and 46.90%, respectively; after screening, Example 2 retained 23 molecular formulas, of which the number of low AI_mod and high H / C molecular formulas was 0. The above comparison shows that the comparative examples that did not use or relaxed the screening conditions retained a large number of low aromaticity characteristic molecules, while Example 2 only retained molecules that met the condition of AI_mod ≥ 0.5.
[0355] The above comparative test results verify the role of the four-dimensional Boolean logic screening and parameter boundary limitation of the present invention. The complex organic matter data detected by high-resolution mass spectrometry contains a large number of non-target molecules. If the multi-dimensional parameters are not limited (Comparative Example 1) or the AI_mod restriction is removed and the H / C range is relaxed (Comparative Example 2), the output results will contain interfering data that do not possess precursor structures. Example 2 of the present invention limits the processing flow to four-dimensional Boolean logic conditions such as AI_mod ≥ 0.5 and H / C between 0.7 and 1.5. The results of Example 2 show that this combination of conditions excludes a high proportion of low-aromatic substances in the background data. This indicates that limiting the multi-dimensional screening parameters and the corresponding numerical range can effectively achieve accurate identification of target precursor molecules.
[0356] Test Example 5:
[0357] 1. Extract the mass spectrometry characteristic peak set obtained and molecular formula identified by Fourier transform ion cyclotron resonance mass spectrometry in Example 2, and the mass spectrometry characteristic peak set obtained and molecular formula identified by high performance liquid chromatography-quadrupole time-of-flight mass spectrometry in Comparative Example 3, and construct PMD networks respectively.
[0358] The PMD network constructed in this stage belongs to a different data processing stage than the PMD network constructed in Test Example 6 after extracting the intersection of EOM, IOM, and AOM.
[0359] 2. Input the mass spectrometry characteristic peak data of the two sets of molecular formula identification into the mass spectrometry data analysis software, use the same preset PMD conversion type, and set the precise mass difference matching tolerance to 1.0 ppm, and construct the PMD network topology diagram respectively.
[0360] 3. Extract network structure parameters from the generated PMD network topology diagram, count the number of connection edges that uniquely match the preset PMD poor quality type, and the number of conflicting connection edges that simultaneously match two or more preset PMD poor quality types.
[0361] Among them, conflicting connection edges refer to connection edges that simultaneously match two or more preset PMD conversion types within the set quality difference matching tolerance for the same precise quality difference.
[0362] 4. Calculate the average node degree of each network topology graph. The average node degree is calculated by dividing twice the sum of the number of all connected edges in the network by the total number of network nodes participating in the statistics. In Example 2 and Comparative Example 3, the number of PMD network nodes participating in the average node degree statistics are 750 and 1000, respectively.
[0363] Record the mass error between the measured accurate mass of each mass spectrometry characteristic peak before PMD matching and the theoretical accurate mass of the derived molecular formula, and calculate the average mass error.
[0364] 5. Summarize the average quality error of the above topology parameters before matching with PMD, and record the number of uniquely matched edges, the number of conflicting edges, and the average node degree of the two networks.
[0365] Table 5. Statistical table of topology parameters for the PMD network constructed in Example 2 and Comparative Example 3.
[0366] Group Average quality error (ppm) Number of uniquely matching connected edges (number of edges) Number of conflicting edges (number of edges) Average node degree Example 2 0.43 890 10 2.4 Comparative Example 3 4.82 753 547 2.6
[0367] Combining the data in Table 5 with Figure 5 The statistical results show that the mass error of the mass spectrometer directly alters the matching composition of the PMD network connections; for example... Figure 5 As shown in (a), the average quality error of Comparative Example 3 was 4.82 ppm, and the average quality error of Example 2 was 0.43 ppm; when a uniform quality difference matching tolerance of 1.0 ppm was set for network construction, as... Figure 5 (b) and Table 5 show that the number of conflicting edges in the network of Comparative Example 3 reached 547, accounting for 42.08% of the total number of edges; the number of conflicting edges in the network of Example 2 was 10, accounting for 1.11%, and the proportion of unique matching edges reached 98.89%; at the same time, the average node degree of Comparative Example 3 was higher than that of Example 2; the data shows that when the quality error of the original data is greater than the set matching tolerance, a high proportion of transformation type conflicts will occur during the network construction process.
[0368] The above comparison illustrates the necessity of using Fourier transform ion cyclotron resonance mass spectrometry (FT-ICRMS) and setting a matching tolerance of 1.0 ppm in this invention. In PMD network analysis, the calculation of node proportions depends on the accurate attribution of connection edges. If the error level of the original data is too high (as in Comparative Example 3), the same measured mass difference will simultaneously conform to multiple theoretical reaction types within the 1.0 ppm matching tolerance, generating a large number of conflicting connection edges. This will lead to deviations in the subsequent statistical results for chlorination or decarboxylation conversion nodes. Example 2 of this invention uses FT-ICRMS to control the measurement mass error within the set matching tolerance of 1.0 ppm. The data from Example 2 shows that the instrument's accuracy conditions effectively reduce multiple matching of conversion types, enabling network connection edges to have a single attribution, thereby ensuring the accuracy of subsequent logical screening based on node proportions.
[0369] Test Example 6:
[0370] 1. Extract the candidate precursor molecular formula list output by parallel determination of extracellular organic matter, intracellular organic matter and mixed organic matter in Example 2 and the candidate precursor molecular formula list output based on a single mixed organic matter component in Comparative Example 4, as the comparison objects of the two data processing flows.
[0371] In Example 2, the number of effective nodes in the PMD network corresponds to the statistical results of the PMD network constructed after extracting the intersection molecular formulas of EOM, IOM, and AOM.
[0372] The effective number of nodes in the PMD network in Comparative Example 4 corresponds to the statistical results of the PMD network directly constructed from the mixed AOM-2 single-component data.
[0373] 2. Molecular formula identification software was used to classify and analyze the molecular formulas in the two output lists by element, and the number of molecular formulas containing sulfur in each list was screened and counted. Among them, CHOS-type molecular formulas were included in the statistics as one of the sulfur-containing molecular formula categories.
[0374] 3. Based on the PMD network topology, count the number of other conversion nodes in the two networks that do not match the set decarboxylation or chlorination quality difference conversion relationship.
[0375] 4. Summarize the total number of detected molecular formulas, the number of sulfur-containing molecular formulas, and the number of other conversion nodes in the two lists, and calculate the proportion of sulfur-containing molecular formulas to the total number of detected molecular formulas, and the proportion of other conversion nodes to the corresponding effective nodes in the PMD network.
[0376] 5. Compare the total number of detected molecular formulas, the number of sulfur-containing molecular formulas and their proportions, and the differences in the distribution of other conversion nodes between the two lists.
[0377] Table 6. Statistics of sulfur-containing molecular formulas and other transformation nodes in the precursor lists of Example 2 and Comparative Example 4.
[0378] Group Total number of molecular formulas detected. Number of sulfur-containing molecular formulas (individuals) Number of valid nodes in the PMD network (number) Number of other conversion nodes (units) Percentage of sulfur-containing molecular formulas (%) Percentage of other conversion nodes (%) Example 2 23 0 358 0 0 0 Comparative Example 4 294 127 346 83 43.20 23.99
[0379] Table 6 and Figure 6 This reflects the molecular formula and node statistics under different data source processing methods; such as Figure 6 As shown in (a) and Table 6, Comparative Example 4, which was treated based on a single mixed organic matter component, output a total of 294 candidate molecular formulas, including 127 sulfur-containing molecular formulas and 83 other transformation nodes; combined with Figure 6 (b) The percentage data shows that in Comparative Example 4, sulfur-containing molecular formulas accounted for 43.20% of the total detected, and other conversion nodes accounted for 23.99% of the effective nodes in the network; Example 2 extracted the intersection molecular formulas based on the parallel systems of EOM, IOM and AOM, and the total number of candidate molecular formulas output was 23, and the number of sulfur-containing molecular formulas and other conversion nodes was 0; it can be seen from the comparison that Example 2 did not detect sulfur-containing molecular formulas and other conversion nodes compared with Comparative Example 4.
[0380] The above data differences illustrate the basis for the invention's setting of extracting the intersection molecular formula of multiple components before network construction; algal organic matter samples contain a large number of non-target background substances; as shown in Comparative Example 4, when network calculations are performed directly on single mixed component data, sulfur-containing molecules and irrelevant transformation nodes that do not participate in the target reaction in the original data are retained and participate in subsequent calculations; Example 2 limits the step of extracting the molecular formula from three sets of data, EOM, IOM and AOM, at the front end of the process; the data in Table 6 show that this step removes background molecules before network construction, reducing the impact of non-target nodes on network topology analysis and subsequent parameter screening.
[0381] Test Example 7:
[0382] 1. Take the high-DOC mother liquor of extracellular organic matter, intracellular organic matter and mixed organic matter from the same batch, which is retained during the parallel scale-up preparation process according to Preparation Example 1 and is calculated according to the required volume based on the target DOC concentration, and establish chlorination reaction systems for each.
[0383] Another sample of equal volume from the same batch was taken as a control before chlorination, and all samples were diluted to 100 mL with ultrapure water to ensure that the DOC concentration of each system was 5.0 mg / L.
[0384] Sodium bromide and sodium iodide were added to the chlorination reaction systems of EOM, IOM, and AOM, and to the corresponding pre-chlorination control samples, respectively, to induce Br. - with I - The concentration was 200 μg / L, and the initial pH was adjusted to 3.0 using sodium hydroxide or hydrochloric acid.
[0385] Subsequently, sodium hypochlorite was added to the EOM, IOM and AOM chlorination reaction systems to make the effective chlorine dosage 11.0 mg / L. No sodium hypochlorite was added to the control sample before chlorination. Each chlorination reaction system was reacted at 25°C in the dark for 48 h.
[0386] After the reaction was completed, the remaining oxidizing halogen equivalent (calculated as Cl2) was determined according to the method in step S1 of Example 1, and sodium thiosulfate solution was added at 1.5 times the theoretical amount of sodium thiosulfate required for complete reaction to quench the reaction.
[0387] 2. The quenched reaction solutions and the control samples before chlorination were treated using the same pretreatment and solid-phase extraction methods.
[0388] Subsequently, ultra-high performance liquid chromatography-tandem mass spectrometry was used to quantitatively detect tetrachlorobenzoquinone, 2,6-dibromobenzoquinone (2,6-DBBQ) and 2,6-diiodobenzoquinone (2,6-DIBQ) in the sample, and the net concentration of each halogenated benzoquinone (HBQs) was recorded (net concentration = concentration detected after chlorination - background concentration of the control before chlorination).
[0389] 3. Fourier transform ion cyclotron resonance mass spectrometry was used to measure the components before and after chlorination. The continuous injection flow rate was set to 120 μL / h, the capillary inlet voltage was 4.0 kV, the ion accumulation time was 0.1 s, the m / z acquisition range was 150~1000, and the number of time domain signal superpositions was 128.
[0390] External standard calibration was performed using 10 mmol / L sodium formate solution, with the mass error controlled within 1.0 ppm. SRNOM was used as the quality control reference sample for verification. After molecular formula identification, the H / C, O / C, DBE, AI_mod, NOSC, and GFE parameters of each molecule were calculated according to unified data processing rules, and the corresponding weighted parameters (H / C_w, O / C_w, DBE_w, GFE_w, and AI_mod_w) were calculated.
[0391] Spearman's rank correlation analysis results were obtained from 15 sets of data (EOM, IOM, and AOM) from Examples 1 to 5, and the correlation between each weighted molecular parameter and the net HBQs generation concentration was recorded (the data of Test Example 7 itself were not included in the correlation calculation).
[0392] Meanwhile, the relative abundance of lignin sample molecular formulas was statistically analyzed based on the VanKrevelen region division rule.
[0393] 4. Independent PMD networks were constructed based on the sets of molecular formulas detected before and after chlorination of EOM, IOM and AOM respectively, and the independent PMD networks were only statistically analyzed at the system level.
[0394] The number of molecular formulas before and after chlorination of EOM, IOM, and AOM were counted separately, and the ratio of the number of molecular formulas detected after chlorination to the number of molecular formulas detected before chlorination was calculated. The proportion of chlorinated poor quality conversion nodes in each network group and the frequency of decarboxylated poor quality conversion matching in the AOM network were also counted.
[0395] The mass spectrometry characteristic peak nodes that participate in at least one preset PMD mass difference conversion relationship are counted as valid nodes of the PMD network. In each independent PMD network, when the same molecular formula corresponds to multiple mass spectrometry characteristic peak records, one mass spectrometry characteristic peak record is retained according to the principle of the maximum mass spectrometry response intensity in the two sets of data before and after chlorination of the system.
[0396] 5. Extract 188 common molecular formulas from the sets detected by EOM, IOM and AOM before and after chlorination. Based on the principle of maximizing the mass spectrometry response intensity in the three systems of EOM, IOM and AOM and the six sets of data before and after chlorination, determine the unique mass spectrometry characteristic peak record corresponding to each common molecular formula, and use it as the input node to construct a common PMD network.
[0397] The shared PMD network is reconstructed independently based on the deduplicated mass spectrometry characteristic peaks, rather than using the intersection of the effective nodes of the independent PMD networks as the set. The network is judged according to the rules of step S6 in Example 1. After the judgment conditions are met, the molecular formula and molecular parameter records corresponding to the mass spectrometry characteristic peak nodes in the shared PMD network are read.
[0398] Finally, the aromaticity index, oxygen-to-carbon ratio, hydrogen-to-carbon ratio, and carbon atom number four-dimensional Boolean logic conditions for the improved molecular formula application are used to screen and output a list of candidate precursor molecular formulas and perform category classification statistics.
[0399] Table 7. Statistics on net HBQs generation concentration, chlorination characteristics and network topology parameters of each organic matter component under the baseline conditions.
[0400] Group TCBQ (μg / L) 2,6-DBBQ (μg / L) 2,6-DIBQ (μg / L) Number of molecular formulas before chlorination Number of valid nodes in the PMD network (number) The ratio of molecular formulas before and after chlorination Extracellular organic matter (EOM) 1.38 1.60 1.00 5091 158 0.042 Intracellular organic matter (IOM) ND ND ND 3899 15 0.053 Mixed organic matter AOM 4.63 5.25 3.77 2342 16 0.186
[0401] Note: In the table above, ND indicates that the concentration of the target analyte in the chlorinated sample is lower than the detection limit of the UPLC-MS / MS method used in this test, and the corresponding net concentration is recorded as ND;
[0402] The method detection limits for TCBQ, 2,6-DBBQ, and 2,6-DIBQ were 0.02 μg / L, 0.01 μg / L, and 0.02 μg / L, respectively.
[0403] Table 7 and Figure 7The data in (a) reflect the concentration differences of haloquinones (HBQs) generated after chlorination of different organic components; among them, the mixed organic matter (AOM) system generated the highest net concentrations of TCBQ, 2,6-DBBQ and 2,6-DIBQ, which were 4.63 μg / L, 5.25 μg / L and 3.77 μg / L, respectively; the net concentrations of the three products in the extracellular organic matter (EOM) system ranged from 1.00 to 1.60 μg / L; and the measured values of the intracellular organic matter (IOM) system were all below the method detection limit.
[0404] Based on this, a PMD network was constructed by extracting the common molecular formulas of the three systems, and after screening with four-dimensional parameters, 12 candidate precursor molecular formulas were finally output (specific parameters are shown in Table 8); Figure 7 (b) It can be seen that all 12 candidate molecular formulas are distributed in the lignin-like region, with hydrogen-carbon ratio (H / C) ranging from 0.71 to 1.08 and oxygen-carbon ratio (O / C) ranging from 0.20 to 0.56. Meanwhile, the data in Table 8 show that their improved aromaticity index (AI_mod) all meet the setting of greater than or equal to 0.5.
[0405] The above test results illustrate the design basis and practical function of the multi-level data processing flow of this invention. Figure 7 Quantitative analysis in (a) demonstrates that AOM and EOM are effective components for generating HBQs, while IOM does not produce such byproducts. This invention sets up these three parallel systems and extracts the intersection molecular formulas, eliminating non-target background data unique to each system in the initial stage through component comparison; combined with Table 8 and... Figure 7 (b) The final screening results show that after the original massive data was transformed and matched by a shared PMD network and limited by four-dimensional Boolean logic parameters, the 12 molecular formulas retained have consistent low hydrogen-to-carbon ratio and high aromaticity characteristics. This result indicates that by combining multi-system intersection extraction, network topology determination and multi-dimensional parameter boundary limitation, non-target interference molecules can be filtered out step by step, and objective extraction of precursors with specific structures can be achieved.
[0406] Table 8. Molecular formulas and key parameters of HBQ candidate precursors.
[0407] Original candidate number Molecular formula Number of carbon atoms (C) Hydrogen to carbon ratio (H / C) Oxygen-to-carbon ratio (O / C) Improve the aromaticity index (AI_mod) 1 <![CDATA[C9H9O5N1]]> 9 1.00 0.56 0.545 7 <![CDATA[C 11 H 10 O4]]> 11 0.91 0.36 0.556 9 <![CDATA[C 14 H 10 O3]]> 14 0.71 0.21 0.680 10 <![CDATA[C 13 H 10 O3]]> 13 0.77 0.23 0.652 11 <![CDATA[C 13 H 10 O4]]> 13 0.77 0.31 0.636 12 <![CDATA[C 15 H 12 O3]]> 15 0.80 0.20 0.630 13 <![CDATA[C 12 H 10 O3]]> 12 0.83 0.25 0.619 15 <![CDATA[C 12 H 12 O4]]> 12 1.00 0.33 0.500 16 <![CDATA[C 13 H 12 O4]]> 13 0.92 0.31 0.545 17 <![CDATA[C 11 H 10 O5]]> 11 0.91 0.45 0.529 20 <![CDATA[C 14 H 12 O3]]> 14 0.86 0.21 0.600 21 <![CDATA[C 12 H 13 O3N1]]> 12 1.08 0.25 0.526
[0408] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A method for identifying HBQ precursors based on FT-ICRMS combined with PMD molecular network, characterized in that, include: S1. Establish a chlorination reaction system for extracellular organic matter (EOM), intracellular organic matter (IOM), and mixed organic matter (AOM) of algae, and a control sample before chlorination. Sodium hypochlorite is added to the chlorination reaction system, and the reaction is quenched afterward. S2, filtration, acidification, and desalting and enrichment via solid-phase extraction; S3, UPLC-MS / MS targeted quantification of the HBQs and calculation of net generated concentration, and FT-ICRMS non-targeted high-resolution determination; S4. Molecular formula identification and molecular parameter calculation; S5. Extract the molecular formulas that are detected together before and after chlorination of the EOM, IOM and AOM systems, and extract the intersection molecular formulas that coexist in the three systems; S6. When the same intersection molecular formula corresponds to multiple mass spectrometry characteristic peak records, one mass spectrometry characteristic peak record shall be retained according to the principle of the maximum mass spectrometry response intensity in the three systems of EOM, IOM, and AOM and the six sets of data before and after chlorination. The mass spectrometry characteristic peak retained after deduplication shall be used as the PMD network node. The measured precise quality difference between the nodes is matched with the absolute value of the theoretical precise quality difference corresponding to the preset PMD conversion type, and the PMD network is constructed using the conversion relationship corresponding to the quality difference as the connection edge. At least one mass spectrometry characteristic peak node that participates in a preset PMD mass difference conversion relationship is a valid node in the PMD network. When the percentage of effective nodes participating in the +Cl-H, +Cl-N, or +Cl+OH mass difference conversion is ≥40% of the total number of effective nodes in the PMD network, and the number of connections matching the -CO2 mass difference conversion is ≥3, the effective nodes of the PMD network are screened according to AI_mod ≥0.5, O / C 0.2~0.6, H / C 0.7~1.5, and the number of carbon atoms 6~20, and the candidate precursor molecular formula is output.
2. The method for identifying HBQ precursors based on FT-ICRMS combined with PMD molecular network according to claim 1, characterized in that, In S1, sodium bromide and sodium iodide are added to both the chlorination reaction system and the pre-chlorination control sample. After the reaction is complete, the remaining oxidizing halogen equivalent is measured, and sodium thiosulfate solution is added to quench the reaction, at a rate of 1.5 times the theoretical amount of sodium thiosulfate required for complete reaction based on the remaining oxidizing halogen equivalent.
3. The method of claim 1, wherein the FT-ICRMS combined with PMD molecular network is used to identify the HBQs precursor. In S2, hydrochloric acid is used to adjust the pH to 2; A styrene-divinylbenzene polymer solid-phase extraction column was used. After activation with methanol and equilibration with pH=2 hydrochloric acid solution, the sample was loaded and eluted with 3 column volumes of the aforementioned pH=2 hydrochloric acid solution. After drying with nitrogen, the sample was eluted with the aforementioned methanol, concentrated by nitrogen blowing, and then reconstituted with a methanol-water mixture at a volume ratio of 50:
50.
4. The method of claim 1, wherein the FT-ICR MS combined with PMD molecular network is used to identify the HBQs precursor. In S3, the target HBQs include TCBQ, 2,6-DBBQ and 2,6-DIBQ; The UPLC-MS / MS uses negative ion multiple reaction monitoring. The FT-ICRMS uses a negative ion electrospray ionization mode, is calibrated with 10 mmol / L sodium formate and verified with SRNOM, and has a mass error of less than 1 ppm.
5. The method of claim 1, wherein the FT-ICR MS combined with PMD molecular network is used to identify the HBQs precursor. In S4, the elemental composition ranges from C1 to 50, H1 to 100, O1 to 30, NO to 5, SO to 2, and P to 1, with an H / C ratio of 0.3 to 2.5 and an O / C ratio of 0 to 1.2, according to [MH]. - The molecular formula was identified using a mass error of 1.0 ppm, a signal-to-noise ratio threshold of 6, and nitrogen rules.
6. The method of claim 1, wherein the FT-ICR MS-based PMD molecular network is used to identify the precursor of HBQs. In step S4, the double bond equivalent (DBE), the modified aromaticity index (AI_mod), the nominal oxidation state of carbon (NOSC), and the standard Gibbs free energy (GFE) of the organic carbon oxidation half-reaction (kJ / molC) are calculated based on the molecular formula obtained from the identification. The calculation formula is as follows: ; ; ; ; In the formula, C, H, O, N, S and P represent the number of carbon, hydrogen, oxygen, nitrogen, sulfur and phosphorus atoms in the corresponding molecular formula, respectively; When the element phosphorus is not present in the molecular formula, the value of P is 0; When C−0.5O−S−N−P=0, the corresponding molecular formula is not included in the calculation of AI_mod, nor is it included in the four-dimensional Boolean logic screening that includes the AI_mod constraint.
7. The method of claim 1, wherein the FT-ICR MS-based PMD molecular network is used to identify the precursor of HBQs. In step S6, the preset PMD conversion types include mass differences corresponding to chlorination reactions (+Cl-H), mass differences corresponding to +Cl-N type mass differences (+Cl-N), mass differences corresponding to chlorination hydroxylation reactions (+Cl+OH), mass differences corresponding to oxidation reactions (+O, +2O, +3O), mass differences corresponding to decarboxylation reactions (-CO2), mass differences corresponding to deoxygenation reactions (-O, -O2), and mass differences corresponding to sulfur-containing reactions (-SO, -SH2). According to the quality difference matching tolerance of 1.0ppm, the measured accurate quality difference between the nodes is matched with the absolute value of the theoretical accurate quality difference corresponding to each preset PMD conversion type; "+" and "-" indicate the direction of change in the element composition corresponding to the PMD conversion type, and do not indicate the sign of the measured precise quality difference between the nodes.
8. The method for identifying HBQ precursors based on FT-ICRMS combined with PMD molecular network according to claim 7, characterized in that, In step S6, the preset PMD conversion type is matched based on the measured precise mass difference between the mass spectrometry characteristic peaks, and the determination of the chlorination mass difference type is not based on whether the molecular formula corresponding to the PMD network node contains the Cl element.
9. The method for identifying HBQ precursors based on FT-ICRMS combined with PMD molecular network according to claim 8, characterized in that, In step S6, for any two PMD network nodes, the measured accurate quality difference between the nodes is the measured accurate quality of the high mass-to-charge ratio node minus the measured accurate quality of the low mass-to-charge ratio node.
10. The method of claim 9, wherein the FT-ICR MS-based PMD molecular network is used to identify the precursor of the HBQs. The candidate precursor molecular formulas are classified according to the Van Krevelen region: The lipid-like ratio is 1.5 ≤ H / C ≤ 2.2 and 0 ≤ O / C < 0.3; The protein / amino acid ratio was 1.5 ≤ H / C ≤ 2.2 and 0.3 ≤ O / C < 0.
67. The lignin sample has a lignin content of 0.7 ≤ H / C < 1.5 and 0.1 ≤ O / C < 0.67; The tannin sample should have a concentration of 0.5 ≤ H / C < 1.5 and a concentration of 0.67 ≤ O / C ≤ 1.
2. The polycyclic aromatic compounds have a ratio of 0.2 ≤ H / C < 0.7 and 0 ≤ O / C < 0.67.
