Tracing method for organophosphorus pollution of water body

Abstract

The invention discloses a method for tracing organophosphorus pollution of a water body, which comprises the following steps of: investigating and surveying the water body, collecting a sample, analyzing mass spectrum, determining an enzymolysis rate coefficient, constructing an APCS-MLR model, and fundamentally correcting the model by utilizing the enzymolysis rate coefficient. The water body organophosphorus pollution traceability technology changes from molecular abundance analysis to ecological risk quantification, the goodness of fit between a model result and an on-site pollution source investigation result is greatly improved, and the problem of treatment resource mismatching caused by neglect of bio-availability difference in a traditional method is thoroughly solved. According to the method, the enzymolysis rate coefficient alpha is introduced, the ecological risk contribution proportion of organic phosphorus input by different pollution sources is accurately quantified, and the problem of source analysis distortion caused by neglecting of DOP degradation characteristics in a traditional method is solved.

Description

Technical Field

[0001] This invention belongs to the field of organophosphorus molecule characterization technology, specifically relating to a method for tracing the source of organophosphorus pollution in water bodies. Background Technology

[0002] Lakes and rivers are important units of the terrestrial surface system and vital carriers of surface water resources, playing a crucial role in maintaining normal human production and life. However, intensified urbanization and population growth, along with the input of large amounts of phosphorus-containing pollutants, have exacerbated the nutrient load on lakes, leading to severe eutrophication problems. Organophosphates (DOPs) are key components of phosphorus and organic matter in water bodies. They originate from diverse and complex pathways, and different forms of DOP can interconvert, making it difficult to analyze their forms and trace pollution sources.

[0003] Traditional methods rely on chemical extraction (such as the phosphomolybdic blue method) to indirectly estimate DOP content at the macroscopic level, but they cannot resolve its chemical structure at all; advanced characterization techniques (such as...) 31 While P-NMR and HPLC-ICP-MS can partially trace the source components, they encounter fundamental bottlenecks due to the general lack of chromophores / fluorophores in DOP compounds. Optical methods such as UV-Vis spectroscopy have a false negative rate exceeding 60% for non-colorimetric compounds, resulting in severely incomplete identification of characteristic molecules of pollution sources. In recent years, Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR-MS), with its ultra-high resolution and precise mass determination capabilities, has become a powerful tool for identifying soluble organophosphates (DOP) at the molecular level. This technology can resolve thousands of mass spectral peaks, assigning a unique molecular formula (containing combinations of elements such as C, H, O, N, and P) to each peak, and revealing key information such as organic matter type, degree of unsaturation, aromaticity, and bioavailability, theoretically offering a possibility for solving the quantitative source tracing problem of complex DOP sources.

[0004] However, the existing source tracing technology system still has significant structural defects in the actual analysis of DOP pollution in water bodies, making accurate source tracing the core challenge of eutrophication control: (1) At the level of technology gap and trace detection: traditional methods such as chemical extraction or conventional spectroscopic / chromatographic techniques can only obtain total data but cannot analyze the structure, or there are detection blind spots and it is difficult to quantify because DOP compounds generally lack chromophores / fluorophores. Although FT-ICR MS can provide molecular-level insights, the existing scheme directly applies the broad-spectrum characterization framework of dissolved organic matter (DOM) and fails to optimize parameters for trace characteristics of DOP in water bodies, resulting in the weak signals of phosphorus-containing molecules being submerged by high background organic matrix noise, and a large number of characteristic fingerprint molecules with source indication significance being missed, ultimately resulting in incomplete molecular maps of pollution sources. (2) At the level of molecular co-elution and model failure: DOP molecules from different pollution sources often co-elute with high overlap in the critical mass-to-charge ratio (m / z) range; existing receptor models (PCA, PMF, etc.) rely on molecular abundance or total concentration for source apportionment, completely ignoring the differences in DOP bioavailability, such as the enzymatic hydrolysis rate of easily degradable phosphate monoesters (such as C6H8O6NP) in domestic sewage (>80%), which are rapidly converted into bioavailable phosphorus; and the stable flame retardants (such as C12H) in industrial wastewater (… 24 Cl3O4P) has an enzymatic hydrolysis rate of <20%, resulting in long-term retention in water bodies.

[0005] Therefore, it is necessary to provide a method for accurately quantifying the source tracing of organic phosphorus pollution in water bodies, and to solve the problem of source apportionment distortion caused by ignoring the degradation characteristics of DOP. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a source tracing method for organic phosphorus pollution in water bodies. By introducing an enzymatic hydrolysis rate coefficient α (a bioavailability weighting factor), the ecological risk contribution ratio of organic phosphorus input from different pollution sources is accurately quantified, thus solving the source apportionment distortion problem caused by neglecting the degradation characteristics of DOP in traditional methods.

[0007] To achieve the above objectives, the present invention provides a method for tracing the source of organic phosphorus pollution in water bodies, comprising the following steps: (1) Analysis of the target water body: Determine the main pollution source types of the target water body and divide the area; (2) Sample collection: Collect water samples from the target area, filter them, and store them in the dark as test samples; (3) Mass spectrometry analysis: after acidification, solid-phase extraction of water samples was performed, and FT-ICR MS was used to obtain mass spectrometry data; (4) Mass spectrometry data analysis: Characteristic molecular formulas are obtained by molecular formula matching calculation; (5) Determination of enzymatic hydrolysis rate coefficient: After acidification of the water sample, alkaline phosphatase was added, the sample was filtered, the DOP concentration was measured, a concentration-time curve was plotted, and the residual rate C at time point was calculated.t / C0 and fitted the first-order kinetic equation, then normalized it to obtain the enzymatic hydrolysis rate coefficient α. i ; (6) Construction and correction of APCS-MLR model: relative abundance of characteristic molecular formulas and α i A weighted abundance matrix is ​​generated, and the contribution rate of pollution source m to the common formula i is calculated using APCS: .

[0008] Preferably, the filtration in step (2) is performed using a 0.45 μm filter membrane.

[0009] Preferably, ultrapure formic acid is used for acidification in step (3), and the pH of the water sample is measured to be <2 after acidification.

[0010] Preferably, a PPL column is used for solid-phase extraction in step (3); before use, it is activated sequentially with liquid chromatography-grade formaldehyde and 0.1% formic acid solution at a flow rate of 5 mL / min; after activation, the acidified water sample is passed through the PPL column, and then the column is rinsed with 0.1% formic acid solution at a flow rate of 1 mL / min, and the column is completely dried with nitrogen; organic matter is eluted with 5 mL of chromatographic-grade methanol and stored at -18℃ in the dark for testing.

[0011] Preferably, the FT-ICR MS detection method in step (3) is Fourier transform ion cyclotron resonance mass spectrometry using a 15T supermagnetic conductor and an electrospray ionization source; the solid-phase extracted sample is characterized using FT-ICR MS negative ion mode: the sample is injected at a rate of 250 μL / h, the capillary voltage is 4.5 kV, and after scanning 128 times under the conditions of ion accumulation time of 0.2 s and m / z scan range of 200-800, the sample is added dropwise, and the signal-to-noise ratio >6 is detected as an effective mass spectrometry peak.

[0012] Preferably, the rule for molecular formula matching calculation in step (4) is that the instrument response is ≥10. 6 S / N≥6; element ratio is H / C<2.0, O / C<1.2.

[0013] Preferably, in step (5), the pH of the acidified water sample is <2, and the final concentration of alkaline phosphatase after addition is 0.1 U / mL; the DOP concentration is determined by the molybdenum blue method.

[0014] Preferably, the first-order kinetic equation in step (5) is ln(C0 / C t )=k i ·t(R 2 ≥0.95); the normalization process is α i =k i / k_max, where k_max is the maximum enzymatic hydrolysis rate constant in the sample.

[0015] Preferably, the weighted abundance matrix in step (6) is Weighted Abundanceim = RawAbundanceim × α i .

[0016] Preferably, the APCS calculation in step (6) includes the following steps: S1: Standardization of the weighted gradation matrix; S2: Using a sample with a concentration of 0, the factor score corresponding to a relative abundance of 0 for each molecular formula; S3: Principal Factor Score - Concentration 0 Sample Score = APCS. Multiple linear regression is performed with APCS as the independent variable and the relative abundance of the characteristic fingerprint molecular formula as the dependent variable. The contribution rate of each pollution source is calculated based on the regression coefficient.

[0017] The beneficial effects of this invention are as follows: 1. By specifically capturing and enriching DOP molecules using a zirconium-based nanofiber solid-phase extraction column, the signal-to-noise ratio was significantly improved to >5. Simultaneously, key parameters of FT-ICR MS were specifically optimized, including focusing on the critical mass range of 300-500 Da and extending the ion accumulation time from 50 ms to 200 ms. This combined strategy greatly enhanced the detection capability of DOP molecules, increasing the number of phosphorus-containing characteristic molecules detected to 1523, resulting in a significant improvement in molecular coverage.

[0018] 2. Unique molecular formulas were obtained through mass spectrometry peak analysis, establishing a molecular-level database of DOP components. Through detailed detection of water bodies and multiple potential pollution sources, and multi-source pollution characteristic analysis, specific organophosphorus molecule groups commonly carried by the water bodies and these pollution sources were screened. Utilizing the relatively stable abundance ratios of these characteristic molecule groups within specific pollution sources, a "pollution source fingerprint spectrum" with clear physical meaning was constructed, providing unique molecular-level identifiers for pollution sources.

[0019] 3. By introducing the enzymatic hydrolysis rate coefficient α (a bioavailability weighting factor) to fundamentally modify the APCS-MLR model, a qualitative leap has been achieved in the source tracing technology of aquatic organophosphorus pollution from "molecular abundance analysis" to "ecological risk quantification." In the Tangxun Lake case study, the key breakthrough lies in the significant reduction of recalcitrant organophosphorus molecules (such as C) from industrial sources by the α coefficient (0.2-0.4). 21 H 30The weight of ONP (Optical Non-Degradable Particles) in domestic sewage was increased, while the ecological risk weight of readily degradable molecules (such as C6H7O6NP, α=1.0) in domestic sewage was increased to 42.1%. This modification greatly improved the agreement between the model results and the field pollution source survey, and completely solved the problem of resource misallocation caused by the neglect of differences in bioavailability in traditional methods.

[0020] 4. Regarding the presence of thiophosphates (C) in agricultural runoff 11 H9O2S3P, α=0.8) and nitrogen-containing heterocyclic compounds in urban stormwater (C 12 H7O 10 The co-elution interference of N2P (α=0.67) was successfully addressed by the α-weighted model, which decoupled the mixed signal and further compressed the contribution rate of unknown sources. Compared with the existing technology (CN117368298A), this scheme significantly improves the number of DOP feature molecules detected. Attached Figure Description

[0021] Figure 1 This is a map showing the general situation of Tangxun Lake and the layout of sampling points in step (2) of Example 1.

[0022] Figure 2 The graph shows the relative peak values ​​of the characteristic fingerprint molecular formulas of the sewage samples from the pollution source in step (6) of Example 1. The black vertical axis represents the relative abundance of fishpond water and urban stormwater sewage, while the red vertical axis represents the relative abundance of domestic sewage.

[0023] Figure 3 This is a bar chart showing the contribution rate of each pollution source to the content of characteristic organic phosphorus molecules in step (7) of Example 1.

[0024] Figure 4 The pie chart shows the average contribution rate of each pollution source to organic phosphorus in Tangxun Lake in step (7) of Example 1.

[0025] Figure 5 This is a bar chart showing the contribution rate of each pollution source to the content of characteristic organophosphorus molecules in Comparative Example 1.

[0026] Figure 6 The pie chart shows the average contribution rate of each pollution source to organic phosphorus in Tangxun Lake in Comparative Example 1. Detailed Implementation

[0027] The technical solution of the present invention will be further explained and described below with reference to the accompanying drawings and specific embodiments. It is worth noting that the following embodiments are only preferred embodiments of the present invention and should not be construed as limiting the present invention. The scope of protection of the present invention should be determined by the contents of the claims. Modifications and substitutions made by those skilled in the art to the technical solution of the present invention without creative effort all fall within the scope of protection of the present invention.

[0028] Example 1 Taking Tangxun Lake, an urban lake in Wuhan, as an example, a source tracing study of organophosphorus pollution was conducted. The specific steps are as follows: (1) Analysis of the target water body: Tangxun Lake is a key national lake and the largest urban lake in China. Its water environment quality directly affects the living environment, residents' health, urban landscape, biodiversity, regional economic development and land value. According to data research and field investigation, the pollution in Tangxun Lake is greatly affected by human activities, including domestic wastewater, excessive fertilizer application and runoff, wastewater from livestock, poultry and aquaculture, and urban stormwater sewage. (2) Sample collection: In 2024, 18 lake water samples and 3 sewage samples from pollution sources were collected, including one each of domestic sewage (collected from the municipal sewage outlet near Maple Leaf International School), fishpond water (collected from the fishpond aquaculture area on the northwest side of the outer lake), and urban stormwater sewage (collected within 24 hours after rainfall >10mm). The 18 lake water samples were evenly collected from the inner and outer lakes of Tangxun Lake, and the lake water was collected at a depth of 20cm below the water surface under no rainfall conditions. Among them, there were 10 surface water samples from the outer lake and 8 surface water samples from the inner lake. Figure 1 The above-mentioned pollution sources directly discharged wastewater into Tangxun Lake; 500 mL of water sample was collected from each sampling point, filtered through a 0.45 μm glass fiber filter membrane, and stored in a brown glass bottle as the test water sample; domestic sewage was used as the first variable factor (VF1), fishpond water as the second variable factor (VF2), and urban stormwater sewage as the third variable factor (VF3). (3) Solid phase extraction: Take the water sample to be tested, add 12 drops of ultrapure formic acid to acidify the water sample to pH<2, and use solid phase extraction for extraction and concentration. The specific method is as follows: Use a PPL column, first activate it with 20 mL of high performance liquid chromatography grade methanol and the same volume of acidified Milli-Q water (pH<2) at a flow rate of 5 mL / min; then pass the acidified water sample through the PPL column at 3 mL / min, and then wash the filter cartridge with 20 mL of acidified Milli-Q water (pH<2) to remove salt (flow rate of 1 mL / min). After being completely dried with nitrogen, elute with 5 mL of chromatographic grade methanol into an acid-washed brown glass container, and store it as a mass spectrometry sample at -18℃ in the dark for testing. (4) Mass spectrometry analysis: The mass spectrometry sample obtained in step (3) was dissolved in high performance liquid chromatography grade methanol (volume ratio of 25:1) and injected into the electrospray source at a rate of 250 µL / h using a syringe pump; the entire device was operated in negative ion mode, with a capillary voltage of 4.5 kV, a capillary column voltage of −320V, and an ion accumulation time of 0.2 s, and then transferred to the ion cyclotron resonance mass spectrometer cell with a flight time of 1.1 ms; the mass-to-charge ratios (m / z) detected ranged from 200 to 800, a total of 128 scans were performed, with a step size of 2 M, and finally a high resolution mass spectrum of phosphorus-containing organic compounds was obtained with a signal-to-noise ratio >5; the internal standard method was used for calibration before detection, that is, 0.1 mg / L of triphenyl deuterated phosphate (TPP-d) was added to the sample. 15 (m / z=368.1523), with 250 μg / L fluorobenzoic acid (m / z=199.0242) as the locked quality standard, the quality error after calibration is ≤±0.5ppm; (5) Mass spectrometry results analysis: High-resolution mass spectra of phosphorus-containing organic compounds were calibrated, and then DataAnalysis (Version 5.0) was used for mass spectrum peak identification and molecular formula matching, with the rule being that the instrumental response (IR) ≥ 10. 6 The S / N ratio is ≥6, and the elemental ratios are H / C < 2.0 and O / C < 1.2. Then, for cases where multiple molecular formulas exist for the same mass spectrum peak, a comprehensive analysis is performed, combining molecular formula matching score, mass errors, and isotope peak shape matching degree (M sigma). The analytical method requires that each molecular formula must contain one phosphorus atom (P1), while allowing unlimited amounts of carbon and hydrogen, oxygen in the range of 0-30 atoms, nitrogen in the range of 0-5 atoms, and sulfur in the range of 0-2 atoms. Strict elemental ratio thresholds are required: (H / C) ≤ 1.2, (O / C) ≤ 1.2, (N / C) ≤ 0.5, to exclude chemically infeasible structures. The isotope verification protocol requires an isotope mode matching score (M sigma) exceeding 0.8, and... 13 The relative intensity error of C is less than 20% to ensure accurate molecular recognition; peak filtering standards include a minimum signal-to-noise ratio (S / N) of 6 and an instrument response threshold ≥10. 6 This effectively eliminated low-quality spectral features. For peaks corresponding to multiple molecular formulas, a comprehensive evaluation was conducted, taking into account molecular formula matching score, mass error, and isotope peak shape consistency (M sigma). For ambiguous cases, a hierarchical sorting method was used: first, the smallest mass error; second, the highest M value; and third, the fewest heteroatoms (oxygen atoms > nitrogen atoms > sulfur atoms were given priority). (6) Determination of characteristic fingerprint molecular formulas: Venn in R language was used to identify identical molecular formulas, and the overlapping parts of the lake water sample and the sewage samples from the three pollution sources were extracted as characteristic fingerprint molecular formulas, totaling 8, namely C 11 H9O2S3P, C 12 H7O 10 N2P, C 19 H 25 OP, C 19 H 26 ONP, C 20 H 28 ONP, C 21 H 30 ONP, C6H8O6NP, C8H 19 O4P; The relative peak values ​​of the eight characteristic fingerprint molecular formulas are used to replace the relative abundance. The distribution of peak height and m / z values ​​among samples are compared. Then, the peak height is normalized to the ratio of a single peak to the sum of all peaks, which is taken as its relative abundance. The molecular formula with the highest relative abundance is selected as the characteristic fingerprint molecular formula of each pollution source wastewater sample. (7) Determination of the enzymatic hydrolysis rate coefficient α: Take 50 mL of water sample, add ultrapure formic acid to acidify to pH < 2, add alkaline phosphatase (Sigma P4252) to a final concentration of 0.1 U / mL, place in a constant temperature shaker at 30℃ and 120 rpm for 24 h, and take 5 mL of sample every 2 h as the sample; after filtering the sample through a 0.22 μm filter membrane, determine the residual concentration of DOP by the molybdenum blue method (detection limit 0.01 μg / L), and plot the concentration-time decay curve (with the initial concentration C0 as the reference, calculate the residual rate C at each time point). t / C0); for each characteristic molecular formula, perform a first-order kinetic fit, requiring a goodness-of-fit R. 2 ≥0.95, calculate the enzymatic hydrolysis rate coefficient α. i =k i / k max , where k max The largest k in the sample i Values ​​(Table 1); Table 1 Enzymatic rate coefficients of characteristic fingerprint molecular formulas

[0029] (8) Construction and modification of the APCS-MLR model: S1: Utilizing the relative abundance of characteristic molecular formulas and α i Generate the weighted abundance matrix (Table 2); Table 2 Weighted Abundance Matrix

[0030] S2: Principal component analysis (PCA) was used to standardize the relative abundance data corresponding to each molecular formula, resulting in normalized factor scores. The KMO value was 0.537, and Bartlett's Sphericity test result was 0.00, indicating a statistically significant correlation between the parameters and the suitability of PCA analysis. The first three principal components were considered to have eigenvalues ​​greater than 1, accounting for 95.387% of the total variance. The component loadings of each parameter in the three principal components are shown in Table 3, where factor loadings of 1-0.75, 0.75-0.5, and 0.5-0.2 were defined as "strong," "moderate," and "weak," respectively. S3: Introduce artificial samples with a concentration of 0 (simply add a row of data with a value of 0), calculate the factor scores corresponding to the relative abundance of 0 for the 8 feature fingerprint molecular formulas, as shown in the following formula: ; In the formula The abundance of formula i is homogeneous in all real samples; denoted as the abundance standard deviation of molecular formula i; Then, the principal factor scores of the eight characteristic fingerprint molecular formulas were subtracted from the factor scores of the human samples to obtain the APCS for each sample (18 lake water samples); then, multiple linear regression analysis was performed with APCS as the independent variable and the relative abundance of the characteristic fingerprint molecular formula as the dependent variable (Y): ; Based on regression coefficients ( , The contribution rate of each pollution source is calculated according to the following formula: Use a im ×APCS im The product of these factors represents the contribution of the pollution source to the common molecular formula factor. The contribution rate of pollution source m to the common molecular formula i is given by the following formula:

[0031] In the formula, For contributions from unidentified sources (unknown sources), the contribution rate formula for unidentified sources (unknown sources) is as follows:

[0032] In the formula, The contribution rates of the relative abundance of the common molecular formula i and the pollution source m are given. is the average absolute principal component factor score of all samples with the common molecular formula relative abundance i.

[0033] After analysis in step (5), the DOM molecular formulas in the samples were finally identified: there were 12,198 to 14,381 effective DOM molecular formulas in 18 lake water samples. The P-containing molecular formulas can be mainly divided into: CHONP, CHONSP, CHOSP and CHOP, totaling 311 to 900. The number of DOM molecular formulas in domestic sewage, fishpond water and urban stormwater sewage were 12,806, 14,381 and 14,381, respectively, and the number of DOP molecular formulas were 341, 361 and 536, respectively.

[0034] After analysis in step (6), C, which has the highest relative abundance in domestic sewage, was found to be... 19 H 26 ONP, C 20 H 28 ONP, C 21 H 30 ONP, C 19 H 25 OP is considered a characteristic fingerprint molecular formula of domestic sewage, and C has the highest relative abundance in fishpond water. 11 H9O2S3P is considered a characteristic fingerprint molecular formula for fishpond water, while C6H8O6NP and C8H are relatively abundant in urban stormwater and sewage. 19 O4P is a characteristic fingerprint molecular formula for urban stormwater and sewage. Figure 2 ).

[0035] Table 3 Loading of the component matrix

[0036] Table 3 shows that VF1 (domestic sewage) affects Zscore (C). 19 H 25 OP), Zscore(C 19 H 26 ONP), Zscore(C 20 H 28 ONP), Zscore(C 21 H 30 ONP) showed strong positive loadings, with values ​​of 0.975, 0.996, 0.996, and 0.968 respectively. These four molecular formulas had relatively high abundances in domestic sewage, which can be attributed to the influence of domestic sewage. VF2 (fishpond water) had a positive impact on Zscore (C). 11 H9O2S3P), Zscore(C 12 H7O 10 N2P) showed a strong positive loading, at 0.951 and 0.935 respectively, and the relative abundance of the common molecular formula was also high in fishpond water samples. Therefore, VF2 can be used to represent the pollution generated in fishpond water. VF3 (urban stormwater sewage) had a positive loading on Zscore (C6H8O6NP) and Zscore (C8H19 The positive loadings of O4P were relatively strong, at 0.951 and 0.888 respectively. The relative abundance of the molecular formulas of these two substances was also high in the phosphorus-containing organic matter of urban sewage. Therefore, VF3 can be interpreted as a result of the influence of urban stormwater and sewage.

[0037] The result of source allocation in step (7) is as follows Figure 3 As shown: The effect of domestic sewage (VF1) on Zscore (C) 19 H 25 OP), Zscore(C) 19 H 26 ONP), Zscore(C 20 H 28 ONP), Zscore(C 21 H 30 The contributions of ONP were the largest, at 67.775%, 94.080%, 83.543%, and 53.123%, respectively. Fishpond water (VF2) contributed the most to Zscore (C). 11 H9O2S3P) and Zscore (C 12 H7O 10 The contributions of urban stormwater (VF3) to Zscore(C6H8O6NP) and Zscore(C8H2P) were 91.007% and 62.652%, respectively. 19 The contributions of O4P were 68.686% and 58.843%. Among these, the contribution rate of unexplained variability to each water quality parameter ranged from 0.165% to 22.047%, with an average of 12.468%. Finally, the average contribution rate of pollution sources to each common molecular formula was calculated, such as... Figure 4 As shown, for the pollution of phosphorus-containing organic matter in the lake water, domestic sewage contributed 38.883%, fishpond water accounted for 27.946%, urban stormwater sewage accounted for 20.704%, and the rest were from unknown sources.

[0038] Comparative Example 1 The method and steps are the same as in Example 1, except that steps (7) and (8) S1 are omitted, and the contribution rate of pollution source m to common formula i is calculated using the following formula:

[0039] In the formula, bi represents the contribution from unidentified sources (unknown sources).

[0040] The results are as follows Figure 5-6 Display: Domestic sewage (VF1 pair) Zscore (C) 19 H 26 ONP) has the largest contribution rate, while fishpond water has the largest contribution rate to Zscore (C). 11The contribution rate of H9O2S3P was the largest, while urban stormwater and sewage contributed the most to Zscore (C6H8O6NP). For the pollution of phosphorus-containing organic matter in lake water, biological sewage contributed 33.9%, fishpond water accounted for 26.7%, urban stormwater and sewage accounted for 22.1%, and the rest were unknown sources.

Claims

1. A method for tracing the source of organophosphorus pollution in water bodies, characterized in that: Includes the following steps: (1) Analysis of the target water body: Determine the main pollution source types of the target water body and divide the area; (2) Sample collection: Collect water samples from the target area, filter them, and store them in the dark as test samples; (3) Mass spectrometry analysis: after acidification, solid-phase extraction of water samples was performed, and FT-ICR MS was used to obtain mass spectrometry data; (4) Mass spectrometry data analysis: Characteristic molecular formulas are obtained by molecular formula matching calculation; (5) Determination of enzymatic hydrolysis rate coefficient: After acidification of the water sample, alkaline phosphatase was added, the sample was filtered, the DOP concentration was measured, a concentration-time curve was plotted, and the residual rate C at time point was calculated. t / C0 and fitted the first-order kinetic equation, then normalized it to obtain the enzymatic hydrolysis rate coefficient α. i ; (6) Construction and correction of APCS-MLR model: relative abundance of characteristic molecular formulas and α i A weighted abundance matrix is ​​generated, and the contribution rate of pollution source m to the common formula i is calculated using APCS: 。 2. The method for tracing the source of organophosphorus pollution in water bodies according to claim 1, characterized in that: The filtration in step (2) uses a 0.45μm filter membrane for vacuum filtration.

3. The method for tracing the source of organophosphorus pollution in water bodies according to claim 1, characterized in that: In step (3), ultrapure formic acid is used for acidification, and the pH of the water sample is tested after acidification to be <2.

4. The method for tracing the source of organophosphorus pollution in water bodies according to claim 1, characterized in that: In step (3), a PPL column is used for solid-phase extraction. Before use, the column is activated sequentially with liquid chromatography-grade formaldehyde and 0.1% formic acid solution at a flow rate of 5 mL / min. After activation, the acidified water sample is passed through the PPL column and then rinsed with 0.1% formic acid solution at a flow rate of 1 mL / min. The column is then completely dried with nitrogen. Organic matter is eluted with 5 mL of chromatographic-grade methanol and stored at -18℃ in the dark until analysis.

5. A method for tracing the source of organophosphorus pollution in water bodies according to claim 1, characterized in that: The FT-ICR MS detection method described in step (3) is Fourier transform ion cyclotron resonance mass spectrometry using a 15T supermagnetic conductor and an electrospray ionization source; the solid-phase extracted sample is characterized using FT-ICR MS negative ion mode: the sample is injected at a rate of 250 μL / h, the capillary voltage is 4.5 kV, and after scanning 128 times under the conditions of ion accumulation time of 0.2 s and m / z scan range of 200-800, the sample is added dropwise, and the signal-to-noise ratio >6 is detected as an effective mass spectrometry peak.

6. The method for tracing the source of organophosphorus pollution in water bodies according to claim 1, characterized in that: The rule for molecular formula matching calculation in step (4) is that the instrument response is ≥10. 6 S / N≥6; element ratio is H / C<2.0, O / C<1.

2.

7. The method for tracing the source of organophosphorus pollution in water bodies according to claim 1, characterized in that: The pH of the acidified water sample in step (5) is <2, and the final concentration after adding alkaline phosphatase is 0.1 U / mL; the DOP concentration is determined by the molybdenum blue method.

8. A method for tracing the source of organophosphorus pollution in water bodies according to claim 1, characterized in that: The first-order kinetic equation described in step (5) is ln(C0 / C t ) = kᵢ·t (R²≥0.95); the normalization treatment is α i =k i / k_max, where k_max is the maximum enzymatic hydrolysis rate constant in the sample.

9. A method for tracing the source of organophosphorus pollution in water bodies according to claim 1, characterized in that: The weighted abundance matrix mentioned in step (6) is Weighted Abundanceim = Raw Abundanceim × α i .

10. The method for tracing the source of organophosphorus pollution in water bodies according to claim 1, characterized in that: The APCS calculation in step (6) includes the following steps: S1: Standardization of the weighted gradation matrix; S2: Using a sample with a concentration of 0, the factor score corresponding to a relative abundance of 0 for each molecular formula; S3: Principal Factor Score - Concentration 0 Sample Score = APCS. Multiple linear regression is performed with APCS as the independent variable and the relative abundance of the characteristic fingerprint molecular formula as the dependent variable. The contribution rate of each pollution source is calculated based on the regression coefficient.

Images