Device and method for detecting harmful substances in wine

Abstract

The invention provides a wine harmful substance detection device and a detection method thereof, and belongs to the technical field of substance detection.The device comprises an array construction module used for tracing a perfluorinated compound pollution thread, collecting adsorption-desorption efficiency and transfer quantity parameters of perfluorinated compounds and combining with a pollution suspected retention set to construct a four-dimensional array; the mass spectrometer screening module is used for screening to-be-measured mass spectrometers meeting the measurement requirement standard; the measurement module is used for preprocessing the wine, dividing the preprocessed wine as a measurement sample into n1 * n2 parts, performing n2 times of measurement on the sample based on each mass spectrometer to be measured, and constructing a measurement matrix for each mass spectrometer to be measured, and n1 is the number of the mass spectrometers to be measured; and the result comprehensive analysis module is used for carrying out analysis and comprehensive analysis on the corresponding measurement matrix based on a measurement requirement standard and outputting a final detection result and a pollution link rectification suggestion. The detection precision of the perfluorinated compounds in the wine is improved.

Description

Technical Field

[0001] This invention relates to the field of substance detection technology, and in particular to a device and method for detecting harmful substances in wine. Background Technology

[0002] Perfluorinated compounds (PFCs) are persistent, bioaccumulative environmental pollutants widely found in soil, water bodies, production equipment, and other environmental and production processes. They can easily enter the wine production chain through various pathways and ultimately remain in wine products. These substances pose potential health hazards; therefore, accurately detecting the residual levels of PFCs in wine and tracing the source of contamination is crucial for ensuring wine quality and safety.

[0003] Existing methods for detecting perfluorinated compounds (PFCs) in wine have several shortcomings: For example, pollution source tracing is incomplete, focusing only on a single production stage and failing to cover the entire static and dynamic process of grape cultivation and production, making it difficult to identify key pollution points; residue prediction lacks scientific basis; and the migration and transformation patterns of PFCs between different stages are not considered, leading to significant deviations in the calculation of maximum and minimum possible residues. Current detection methods generally employ fixed mass spectrometry for single measurements without accurately matching the instrument to the characteristics of PFCs and detection requirements, affecting the reliability of the results. All these factors contribute to low detection accuracy. Furthermore, current biochemical sensors used for detecting hazardous substances in food may lack sufficient specificity for identifying PFCs in wine, are susceptible to matrix interference, and are unable to meet the requirements for detecting low concentrations of residues.

[0004] Therefore, the present invention provides a device and method for detecting harmful substances in wine. Summary of the Invention

[0005] This invention provides a detection device and method for harmful substances in wine, which enables accurate detection of perfluorinated compound residues through full-process pollution source tracing, scientific data model construction, precise instrument screening, and optimized pretreatment and data processing procedures, providing technical support for the quality and safety management of wine.

[0006] This invention provides a device for detecting harmful substances in wine, comprising: The array construction module is used to trace the perfluorinated compound (PFCC) contamination threads of grapes in the planting, static production, and dynamic production processes, and to collect the adsorption-desorption efficiency and transfer parameters of PFCCs between each stage. Combined with the suspected contamination retention set of each contamination thread, a four-dimensional array is constructed. The mass spectrometer screening module is used to analyze the maximum and minimum possible residues of perfluorinated compounds in wine based on the four-dimensional array, call the perfluorinated compound carbon chain length-mass spectrometry response specificity database, lock the determination requirement standards, and screen the mass spectrometers to be measured that meet the determination requirement standards from the standard-measurement comparison table. The number of the mass spectrometers to be measured is at least two, and the determination requirement standards include: detection limit and specific response intensity. The measurement module is used to pre-treat the wine and divide the pre-treated wine into n1×n2 portions as measurement samples. Based on each mass spectrometer to be measured, n2 measurements are performed on the samples to construct a measurement matrix for each mass spectrometer to be measured, where n1 is the number of mass spectrometers to be measured. The measurement matrix includes: the types and contents of perfluorinated compounds in each measurement, and is associated with the determination conditions of each measurement. The determination conditions include: chromatographic conditions and mass spectrometry conditions. The results comprehensive analysis module is used to analyze the corresponding measurement matrix based on the measurement requirements standard, and to obtain a single detection result by combining the measurement results of the blank experimental group of the corresponding mass spectrometer to be measured. It also performs comprehensive analysis on all single detection results and outputs the final detection result and suggestions for rectification of pollution links.

[0007] Preferably, the array construction module includes: The first traceability unit is used to trace the perfluorinated compound pollution threads in the grape growing process. It collects data on the residual amount of perfluorinated compounds in the soil, the concentration of perfluorinated compounds in irrigation water, and the fluorinated adjuvant components of agricultural inputs as pollution thread factors in the growing process. The second traceability unit is used to trace the perfluorinated compound contamination threads in the static process of wine production. It collects information on the types of fluorinated contact materials of production equipment, the fluorinated coating information of storage containers, and the detection of perfluorinated compounds in processing auxiliary materials, as factors of the contamination threads in the static process of production. The third traceability unit is used to trace the perfluorinated compound contamination threads in the dynamic process of wine production. It collects data on the contact time of fluorinated equipment during material transfer, the amount of perfluorinated compound released from equipment during ultrasonic processes, and the perfluorinated compound settling rate in the production environment as factors of the contamination threads in the dynamic process of production.

[0008] Preferably, the array construction module further includes: The simulation unit is used to simulate the initial range of theoretical residual concentrations of perfluorinated compounds migrating from the contact medium to the matrix, based on the surface characteristics of the contact medium in each contamination thread, the carbon chain length of the perfluorinated compounds, and the adsorption coefficient of the matrix for the corresponding type of perfluorinated compounds. A deviation calibration unit for calling a preset cross - matrix pollution characteristic matching library of perfluorinated compounds - matrix, matching historical pollution data with the same contact medium type and perfluorinated compound type as the current pollution thread, and performing deviation calibration on the initial interval of the theoretical residual concentration; An experimental unit for conducting customized small - scale verification experiments on the current pollution thread. Specifically, taking the contact conditions in actual production as parameters, performing a simulated contact experiment between the matrix and the contact medium of the corresponding pollution thread, and detecting the actual concentration of perfluorinated compounds in the matrix after the experiment; An interval comparison unit for comparing the actual concentration obtained from the small - scale verification with the calibrated theoretical residual concentration interval to obtain a pollution suspected retention set including perfluorinated compound types, the corrected residual concentration interval, and the pollution contact path; An array construction unit for constructing a four - dimensional array of pollution link - pollution thread factor - cross - link dynamic migration parameter - pollution suspected retention set based on the pollution thread factors in the planting link, production static link, and production dynamic process, and combining the pollution suspected retention sets corresponding to each pollution thread, the adsorption - desorption efficiency and transfer amount parameters of perfluorinated compounds between each link;

[0009] Preferably, the mass spectrometer screening module includes: An interval extraction unit for extracting the corrected residual concentration interval of perfluorinated compounds in the pollution suspected retention set corresponding to each pollution thread factor from the four - dimensional array; A weight acquisition unit for obtaining the contribution weight value of the corresponding pollution thread factor based on the ratio of the product of the contact duration and contact area between the contact medium of the corresponding pollution thread factor and the matrix to the sum of the products of the contact media of all pollution thread factors, and combining the cross - link dynamic parameters; A multiplication unit for multiplying the upper and lower limits of the corrected residual concentration interval corresponding to the corresponding pollution thread factor by the corresponding contribution weight value respectively to obtain the weighted residual concentration interval of the corresponding pollution thread factor, and summing the upper limits of all factor - weighted intervals to obtain the maximum possible residue and summing the lower limits to obtain the minimum possible residue; A residue adjustment unit for calling the actual concentration reference value of the small - scale verification experiment, and adjusting the corresponding residue if the deviation between the possible residue and the actual concentration reference value exceeds the threshold;

[0010] Preferably, the measurement module includes: when the maximum possible residual concentration is u1, the dosage of the extractant is 2 times the volume of the wine sample; When the maximum possible residual concentration < u1, the dosage of the extractant is increased to 3 times the volume of the wine sample, where u1 is the concentration threshold.

[0011] Preferably, the result comprehensive analysis module includes: The vector construction unit is used to extract features from the measurement matrix according to each standard in the measurement requirement standard to obtain the first feature and construct the feature vector. At the same time, it performs external interference analysis on the measurement results of the blank experimental group to obtain the interference vector. A limit processing unit is used to perform limit processing on the feature vector to obtain a limit processing vector. The vector optimization unit is used to perform position difference analysis on each element in the limit processing vector according to each interference factor in the interference vector, determine the posterior distribution of each element, and correct the corresponding elements according to the posterior distribution to obtain the optimized vector, which is regarded as a single detection result.

[0012] Preferably, the result comprehensive analysis module further includes: The coefficient determination unit is used to determine the detection fit coefficient of the perfluorinated compound corresponding to each single detection result based on the carbon chain length and polarity of the target perfluorinated compound in the four-dimensional array. At the same time, it determines the measurement condition fit coefficient based on the chromatographic resolution and mass spectrometry response intensity in the measurement matrix corresponding to each single detection result. The weight calculation unit is used to calculate the confidence weight of a single detection result based on the product of the detection fit coefficient and the measurement condition fit coefficient. The elimination unit is used to calculate the mean and standard deviation of each perfluorinated compound in all single test results. If the absolute value of the difference between the concentration value of any perfluorinated compound and the corresponding mean in the corresponding single test result is greater than twice the standard deviation, and the corresponding confidence weight is less than the preset weight, then the concentration value of the corresponding perfluorinated compound in the corresponding single test result is eliminated. The final calibration unit is used to perform weighted calculations on the concentration values ​​of each retained perfluorinated compound to obtain the initial test results. At the same time, it retrieves the actual concentration reference values ​​of the corresponding perfluorinated compounds in the small-scale experimental verification to calibrate the initial test results, obtain the final test results, and output pollution control suggestions.

[0013] This invention provides a method for detecting harmful substances in wine, comprising: Step 1: Trace the perfluorinated compound (PFCC) contamination threads of grapes in the planting, static production, and dynamic production processes, and collect the adsorption-desorption efficiency and transfer parameters of PFCCs in each process. Combine this with the suspected contamination retention set of each contamination thread to construct a four-dimensional array. Step 2: Based on the analysis of the maximum and minimum possible residues of perfluorinated compounds in wine using the four-dimensional array, call the perfluorinated compound carbon chain length-mass spectrometry response specificity database, lock the determination requirements standard, and select the mass spectrometer to be measured from the standard-measurement comparison table that meets the determination requirements standard. The number of the mass spectrometer to be measured is at least 2, and the determination requirements standard includes: detection limit and specific response intensity. Step 3: Pre-treat the wine and divide the pre-treated wine into n1×n2 portions as measurement samples. Perform n2 measurements on each sample based on the mass spectrometer to be measured, and construct a measurement matrix for each mass spectrometer to be measured, where n1 is the number of mass spectrometers to be measured. The measurement matrix includes: the types and contents of perfluorinated compounds in each measurement, and is associated with the determination conditions for each measurement. The determination conditions include: chromatographic conditions and mass spectrometry conditions. Step 4: Analyze the corresponding measurement matrix based on the aforementioned measurement requirements standards, and obtain a single detection result by combining the measurement results of the blank experimental group of the corresponding mass spectrometer to be measured. Then, conduct a comprehensive analysis of all single detection results and output the final detection result and suggestions for rectifying the pollution process.

[0014] Compared with the prior art, the beneficial effects of this application are as follows: By constructing a four-dimensional array through end-to-end pollution source tracing, comprehensive integration of pollution information is achieved. Mass spectrometry is selected based on compound characteristics and residual amounts to ensure targeted detection. A measurement matrix is ​​constructed through multiple measurements using multiple instruments, combined with blank experiments to eliminate interference, improving the accuracy and reliability of detection results. The final output of detection results and suggestions for rectifying pollution links provides precise guidance for quality control in the wine production process, effectively reducing the risk of perfluorinated compound residues. Furthermore, the pollution source tracing data and preprocessing scheme of this invention can assist biochemical sensors in optimizing detection parameters, reducing matrix interference, and improving the detection stability of perfluorinated compounds.

[0015] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0016] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1This is a structural diagram of a detection device for harmful substances in wine according to an embodiment of the present invention; Figure 2 This is a flowchart of a method for detecting harmful substances in wine according to an embodiment of the present invention. Detailed Implementation

[0018] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0019] This invention provides a device for detecting harmful substances in wine, such as... Figure 1 As shown, it includes: The array construction module is used to trace the perfluorinated compound (PFCC) contamination threads of grapes in the planting, static production, and dynamic production processes, and to collect the adsorption-desorption efficiency and transfer parameters of PFCCs between each stage. Combined with the suspected contamination retention set of each contamination thread, a four-dimensional array is constructed. The mass spectrometer screening module is used to analyze the maximum and minimum possible residues of perfluorinated compounds in wine based on the four-dimensional array, call the perfluorinated compound carbon chain length-mass spectrometry response specificity database, lock the determination requirement standards, and screen the mass spectrometers to be measured that meet the determination requirement standards from the standard-measurement comparison table. The number of the mass spectrometers to be measured is at least two, and the determination requirement standards include: detection limit and specific response intensity. The measurement module is used to pre-treat the wine and divide the pre-treated wine into n1×n2 portions as measurement samples. Based on each mass spectrometer to be measured, n2 measurements are performed on the samples to construct a measurement matrix for each mass spectrometer to be measured, where n1 is the number of mass spectrometers to be measured. The measurement matrix includes: the types and contents of perfluorinated compounds in each measurement, and is associated with the determination conditions of each measurement. The determination conditions include: chromatographic conditions and mass spectrometry conditions. The results comprehensive analysis module is used to analyze the corresponding measurement matrix based on the measurement requirements standard, and to obtain a single detection result by combining the measurement results of the blank experimental group of the corresponding mass spectrometer to be measured. It also performs comprehensive analysis on all single detection results and outputs the final detection result and suggestions for rectification of pollution links.

[0020] In this embodiment, for example, the specific implementation process of measuring the sample based on the mass spectrometer A1 (a triple quadrupole tandem mass spectrometer using an electrospray negative ion source and multiple reaction monitoring mode) is as follows: Sample pretreatment: Take 5 mL of the pretreated wine sample and place it in a polypropylene centrifuge tube. Add 20 μL of isotope internal standard mixture, wherein the concentration of the mixture is 200 ng / mL and the internal standard includes isotope labels with structures similar to the target analyte. Vortex mix for 1 min, add 10 mL of alkaline methanol-ethanol mixed extraction agent, extract by sonication for 30 min, centrifuge at 10000 r / min for 10 min, and collect the supernatant. Purification: Pass the supernatant through the composite purification column, collect the effluent by gravity, add 1.5g of sodium chloride to the effluent, vortex to separate the layers, and take 5mL of the upper organic phase. Concentration and redissolution: The upper organic phase was dried under nitrogen at 40°C, 1 mL of methanol was added for redissolution, and the solution was filtered through a 0.22 μm nylon membrane to obtain the test solution; Separation and detection: The test solution was injected into the separation and detection module, and gradient elution was performed using 2 mmol / L ammonium formate aqueous solution-methanol as the mobile phase at a flow rate of 0.4 mL / min, a column temperature of 35℃, and an injection volume of 2 μL; the mass spectrometer capillary voltage was 2.0 kV, the desolvation gas temperature was 500℃, and the desolvation gas flow rate was 1000 L / h. Comparative analysis: Qualitative analysis was performed by comparing the retention times of chromatographic peaks and the abundance ratio of characteristic ions between the test solution and perfluorinated compound standards, and quantitative analysis of perfluorinated compounds was performed using the internal standard method.

[0021] The gradient elution program was as follows: 0-0.5 min, 80% 2 mmol / L ammonium formate aqueous solution; 0.5~9 min, 80%~5% 2mmol / L ammonium formate aqueous solution; 9.0~12 min, 5% 2mmol / L ammonium formate aqueous solution; 12 min to 12.1 min, restore to 80% 2 mmol / L ammonium formate aqueous solution, and equilibrate for 15 min; Among them, perfluorinated compounds include 23 types with or Perfluoroalkyl compounds with a structure.

[0022] The composite purification column consists of an upper layer of 500mg ENVI-18 packing and a lower layer of 100mg N-propylethylenediamine packing. The separation and detection module includes a chromatographic column and a triple quadrupole tandem mass spectrometer.

[0023] In this embodiment, the adsorption-desorption efficiency refers to the ratio of the amount of perfluorinated compound adsorbed on the surface of the contact medium at different stages to the amount subsequently desorbed from the contact medium into the matrix at the next stage, reflecting the migration ability of perfluorinated compound between the contact medium and the matrix.

[0024] In this embodiment, the transfer amount parameter refers to the specific mass or concentration value of the perfluorinated compound transferred from the upstream matrix / contact medium to the downstream matrix / contact medium between two adjacent production stages.

[0025] In this embodiment, the measurement requirement standard refers to the quantitative indicators set for the detection process to ensure the accuracy and reliability of the detection results, including the detection limit, which is the lowest concentration of perfluorinated compounds that the instrument can accurately detect, and the specific response intensity, which is the intensity of the characteristic ion response signal corresponding to the target perfluorinated compound, which must be able to be effectively distinguished from the response signals of interfering substances.

[0026] In this embodiment, the standard-measurement comparison table refers to a table that records the correspondence between different testing instrument models, performance parameters and various testing requirement standards, which is used to quickly screen testing instruments that meet the requirements.

[0027] In this embodiment, the measurement conditions refer to the key parameters that affect the detection results during mass spectrometry detection, including: Chromatographic conditions: mobile phase composition, flow rate, column temperature, injection volume, etc. Mass spectrometry conditions: ion source type, capillary voltage, desolvation gas temperature and flow rate, etc.

[0028] In this embodiment, parameter acquisition involves laboratory testing and on-site monitoring to collect parameters such as the adsorption-desorption efficiency and transfer amount of perfluorinated compounds at each stage. For example, at the junction of the planting stage and the static production stage, the amount of perfluorinated compounds adsorbed by the grape fruit (substrate) from the soil (contact medium) and the amount desorbed from the fruit into the juice during subsequent washing and crushing processes are collected to calculate the adsorption-desorption efficiency. The transfer amount parameter is determined by detecting the difference between the concentration of perfluorinated compounds in the crushed juice and the concentration in the fruit.

[0029] In this embodiment, the suspected contamination set is determined for each contamination thread by simulating and calibrating the residual concentration range based on factors such as the characteristics of the contact medium and the carbon chain length of the perfluorinated compound, and then verifying the suspected contamination set through small-scale experiments.

[0030] In this embodiment, the four-dimensional array is constructed with the planting stage / static production stage / dynamic production process as the first dimension (contamination stage), the contamination thread factor corresponding to each stage as the second dimension, the adsorption-desorption efficiency and transfer parameters as the third dimension (cross-stage dynamic migration parameters), and the suspected contamination retention set as the fourth dimension, and is integrated to form a four-dimensional array. For example, assuming the pollution thread factors in the planting stage include a soil perfluorooctane sulfonate (PFOS) residue of 10 ng / kg, an irrigation water PFOS concentration of 5 ng / L, and a perfluorohexane sulfonate (PFHxS) content of 20 ng / g in fluorinated adjuvants used in planting materials; in the cross-stage dynamic migration parameters, the soil-grape adsorption-desorption efficiency is 30%, and the transfer amount is 8 ng / kg; the suspected pollution residue set is: {PFOS: residue concentration range 5-12 ng / kg, contact path: soil → grape; PFHxS: residue concentration range 3-8 ng / kg, contact path: agricultural materials → grape}, then the entries corresponding to the planting stage in the four-dimensional array are: planting stage; {soil PFOS residue, irrigation water PFOS concentration, agricultural material PFHxS content}; {adsorption-desorption efficiency 30%, transfer amount 8 ng / kg}; {PFOS: 5-12 ng / kg (soil → grape), PFHxS: 3-8 ng / kg (agricultural materials → grape)}.

[0031] In this embodiment, the residue analysis is based on a constructed four-dimensional array to calculate the maximum and minimum possible residues of perfluorinated compounds in the wine.

[0032] In this embodiment, the database access is a pre-established carbon chain length-mass spectrometry response specificity database. This database records the mass spectrometry response characteristics of perfluorinated compounds with carbon chain lengths of 4-12. For example, the response intensity of PFOS with a carbon chain length of 8 under an electrospray negative ion source is... The detection sensitivity is 2 ng / L; the response intensity of PFHxS with a carbon chain length of 6 is... The detection sensitivity is 3 ng / L.

[0033] The determination of the testing requirements is based on the residue analysis results and the response characteristics in the database. For example, if the maximum possible residue is 15 ng / kg and the minimum possible residue is 3 ng / kg, then the detection limit is set to ≤2 ng / kg, and the specific response intensity is determined accordingly. This ensures that the target material's response signal is significantly higher than the interference signal.

[0034] In this embodiment, instrument selection involves consulting a standard-measurement comparison table, which records the performance parameters of different mass spectrometers. For example, the detection limit of mass spectrometer A (triple quadrupole tandem mass spectrometer) is 1 ng / kg, and its specific response intensity is... The detection limit of mass spectrometer B (time-of-flight mass spectrometer) is 1.5 ng / kg, and its specific response intensity is... The detection limit of mass spectrometer C (ion trap mass spectrometer) is 3 ng / kg, and its specific response intensity is... Screening was conducted to select samples that met the detection limit ≤2ng / kg and specific response intensity. Mass spectrometer A and mass spectrometer B are used as the mass spectrometers to be measured, and the quantity is two, satisfying the requirement of at least two.

[0035] In this embodiment, sample pretreatment involves preprocessing the wine sample, such as extraction, purification, and concentration.

[0036] The sample division is as follows: if the number of mass spectrometers to be measured is n1=2 (mass spectrometer A and mass spectrometer B), and n2=3 (each instrument measures 3 times), then the pretreated wine sample will be divided into 2×3=6 portions, and each portion will have a volume of 5mL.

[0037] Sample measurements were performed using mass spectrometer A and mass spectrometer B, respectively, on three corresponding samples. During the measurements, the types and concentrations of perfluorinated compounds, as well as the corresponding chromatographic and mass spectrometric conditions, were recorded for each measurement. For example, in the first measurement using mass spectrometer A, the PFOS content was detected as 7.2 ng / kg and the PFHxS content as 4.5 ng / kg. The chromatographic conditions were: mobile phase 2 mmol / L ammonium formate aqueous solution-methanol, flow rate 0.4 mL / min, column temperature 35℃, and injection volume 2 μL; the mass spectrometric conditions were: electrospray negative ion source, capillary voltage 2.0 kV, desolvation gas temperature 500℃, and desolvation gas flow rate 1000 L / h.

[0038] Measurement matrix construction: Taking mass spectrometer A as an example, the constructed measurement matrix is ​​shown in Table 1: Table 1 Measurement Matrix of Mass Spectrometer A

[0039] Similarly, construct the measurement matrix for mass spectrometer B.

[0040] In this embodiment, obtaining a single detection result involves extracting features from the measurement matrix of each mass spectrometer to be measured according to the measurement requirements, constructing a feature vector. Simultaneously, for the blank experimental group, using simulated wine samples free of perfluorinated compounds, the measurement results are analyzed using the same preprocessing and detection procedures to identify external interference factors, such as trace amounts of perfluorinated compounds in the laboratory environment and instrument noise, resulting in an interference vector. The feature vector is then subjected to extreme processing, such as taking the extreme range of each measurement value, and combined with the interference vector for positional difference analysis to correct the feature vector and obtain the single detection result. For example, the single detection result for mass spectrometer A is PFOS: 6.9-7.1 ng / kg, PFHxS: 4.4-4.5 ng / kg; the single detection result for mass spectrometer B is PFOS: 7.0-7.2 ng / kg, PFHxS: 4.3-4.6 ng / kg.

[0041] Comprehensive analysis involves combining all individual test results, calculating the weighted average for each perfluorinated compound, removing outliers, and calibrating the results by incorporating actual concentration reference values ​​from small-scale experiments to obtain the final test results. For example, the final test results might be: PFOS: 7.0 ng / kg, PFHxS: 4.5 ng / kg.

[0042] The rectification recommendations are based on pollution source tracing information and test results. If the excessive PFOS residue in the soil during the planting process is the main cause of pollution, the rectification recommendations are: replace the soil in the planting area or use soil remediation technology to reduce PFOS residue; if the fluorine-containing coating of the production equipment is the source of pollution, it is recommended to replace the production equipment with one without a fluorine coating.

[0043] In this embodiment, the experimental conditions were standardized as follows: wine sample volume 5 mL, extractant was alkaline methanol-ethanol (7:3), chromatographic column was C18 column (150 mm × 2.1 mm, 1.8 μm), and the comparative experimental data are shown in Table 2.

[0044] Table 2 Comparative Experiment Data Table

[0045] In this embodiment, the correlation logic between the pollution thread factor and the cross-stage migration parameter in the four-dimensional array is as follows: Taking soil PFOS in the planting stage as an example, when the soil PFOS residue (factor) increases by 10 ng / kg, the experimental verification shows that its influence coefficient on the soil-grape adsorption-desorption efficiency is 0.8, that is, the adsorption-desorption efficiency is improved by 8%; a quantitative correlation formula between the factor value and the migration parameter is established: Migration parameter correction value = initial migration parameter × (1 + 0.08 × (actual factor value - factor baseline value) / factor baseline value), where the factor baseline value is the industry average soil PFOS residue of 8 ng / kg.

[0046] In this embodiment, the linkage mechanism between carbon chain length, mass spectrometry response specificity, and residue level for instrument screening is as follows: when the carbon chain length of a perfluorinated compound is ≥8 (e.g., PFOS) and the maximum possible residue concentration is ≥5 ng / kg, the detection limit is locked at ≤0.8 ng / kg, and the specific response intensity is [not specified]. When the carbon chain length is <6 (e.g., PFPeA) and the maximum possible residual concentration is <2 ng / kg, the determination requirements are adjusted to: detection limit ≤0.3 ng / kg, specific response intensity. This dynamic adjustment rule avoids the problem of missed detection of low-residue short-carbon-chain compounds caused by fixed-standard screening instruments in existing technologies. Through verification with 20 groups of compounds with different carbon chain lengths, the missed detection rate has been reduced from 15% in existing technologies to below 3%.

[0047] In this embodiment, the perfluorinated compound carbon chain length-mass spectrometry response specificity database covers the following: The database covers perfluorinated compounds with carbon chain lengths ranging from 4 to 12, specifically including PFBA (C4), PFPeA (C5), PFHxA (C6), PFHpA (C7), PFOA (C8), PFNA (C9), PFDA (C10), PFUnA (C11), PFDoA (C12), PFBS (C4), PFHxS (C6), and PFOS (C8). Response intensity measurement conditions: Ion source was electrospray negative ion source (ESI-), capillary voltage 2.0kV, desolvation gas temperature 500℃, desolvation gas flow rate 1000L / h, collision energy 15-30eV (adjusted according to carbon chain length: C4-C6 15eV, C7-C9 22eV, C10-C12 30eV); Typical compound response data are shown in Table 3: Table 3 Response data of typical compounds

[0048] In this embodiment, the standard-measurement comparison table includes five commonly used mass spectrometer parameters, as shown in Table 4: Table 4 Standard-Measurement Comparison Table

[0049] In this embodiment, the adsorption-desorption efficiency determination steps are as follows: Adsorption capacity determination: 100g of contact medium (such as soil) is mixed with 100mL of matrix (such as grape juice), shaken at 25℃ for 24h, centrifuged (10000r / min, 10min), and the supernatant is used to detect the concentration of perfluorinated compounds C1 using a mass spectrometer. Adsorption capacity = (total amount of compounds in the initial medium - total amount of compounds in the remaining medium); Desorption capacity determination: The above-mixed medium is centrifuged to separate the media, and the matrix (grape juice) is mixed with a new contact medium (such as a material sheet of production equipment), shaken at 25℃ for 12h, and the concentration of compounds C2 in the new matrix is ​​detected. Desorption capacity = C2 × matrix volume; Efficiency calculation: Adsorption-desorption efficiency = (desorption capacity / adsorption capacity) × 100%, and the average value is taken from 3 parallel determinations.

[0050] Transfer amount parameter determination steps: Upstream matrix concentration detection: Take 10g of upstream matrix (e.g., grape), extract and detect the compound concentration Cup; Downstream matrix concentration detection: Take 10mL of downstream matrix (e.g., grape juice) and detect the compound concentration Cdown; Transfer amount calculation: Transfer amount = (Cdown × total mass of downstream matrix - Cup × total mass of upstream matrix × migration rate), where the migration rate is the proportion of upstream matrix converted into downstream matrix (e.g., grape juice extraction rate 80%).

[0051] In this embodiment, the conditions for the full-process small-scale test are as follows: Small-scale planting conditions: 500g soil sample (PFOS residue 12ng / kg), 3 grape seedlings (20cm tall), 500mL irrigation water (PFOS concentration 6ng / L) per week, 12h / day light, 25℃ temperature, and 30 days of cultivation. Production static pilot test conditions: production equipment material sheet (fluororubber, 30% fluorine content) 10cm×10cm, grape juice 500mL (initially free of perfluorinated compounds), soaking at 25℃ for 24h, stirring speed 100r / min; Production dynamic pilot test conditions: ultrasonic equipment (power 300W), 500mL grape juice, ultrasonic time 20min, temperature 30℃, 3 sedimentation sampling points in the production environment (height 1.5m, spacing 2m), sampling plate (diameter 5cm) coated with C18 adsorbent, sampling for 24h.

[0052] In this embodiment, the Bayesian model parameter calibration method is as follows: Interference influence coefficient calibration: Blank experimental group setting: Take 5 mL of simulated wine sample (composition: ethanol 12%, sucrose 2%, citric acid 0.5%) without perfluorinated compounds, operate according to the preprocessing steps of this application, and detect the interference signal value. (e.g., environmental PFOS interference of 0.1 ng / kg); Standard test: A simulated wine sample containing PFOS standard (concentration 5 ng / kg) was taken, and the standard response value Sstandard was obtained. ; .

[0053] Measurement standard deviation calibration: The same standard sample (PFOS 5 ng / kg) was measured 10 times to obtain the following concentration values: 5.0, 5.1, 4.9, 5.0, 5.2, 4.8, 5.0, 5.1, 4.9, 5.0 ng / kg. The standard deviation was then calculated.

[0054] In this embodiment, the alkaline methanol-ethanol mixed extractant was prepared as follows: 70 mL of methanol (chromatographic grade) + 30 mL of ethanol (chromatographic grade) + 0.5 mL of ammonia (25%, analytical grade), mixed thoroughly, and then ultrasonically degassed for 10 min; the composite purification column specifications were as follows: column length 10 cm, inner diameter 1 cm, upper layer packed with 500 mg ENVI-18 packing material (particle size 50 μm, pore size 80 Å), lower layer packed with 100 mg N-propylethylenediamine (PSA) packing material (particle size 40 μm, pore size 60 Å), with the packing materials separated by quartz wool, and activated with 5 mL of methanol before use; the concentration and reconstitution conditions were as follows: nitrogen blowing at 40℃ (nitrogen flow rate 5 mL / min) until dry, 1 mL of methanol (chromatographic grade) was added, vortexed for 30 s, filtered through a 0.22 μm nylon filter membrane (diameter 13 mm), and the filtrate was stored in the dark (4℃) and analyzed within 24 h.

[0055] In this embodiment, the detection limit refers to the method detection limit, that is, the compound concentration when the signal-to-noise ratio S / N=3; the specific response intensity refers to the ratio of the characteristic ion response signal of the target compound to the interfering ion response signal ≥10.

[0056] The beneficial effects of the above technical solution are as follows: a four-dimensional array is constructed through full-process pollution source tracing to achieve comprehensive integration of pollution information; a mass spectrometer is selected and adapted based on compound characteristics and residual amounts to ensure the targeting of detection; a measurement matrix is ​​constructed through multiple measurements by multiple instruments, and interference is eliminated by combining blank experiments to improve the accuracy and reliability of detection results; the final output of detection results and suggestions for rectification of pollution links provides precise guidance for quality control in the wine production process and effectively reduces the risk of perfluorinated compound residues.

[0057] This invention provides a device for detecting harmful substances in wine, wherein the array construction module includes: The first traceability unit is used to trace the perfluorinated compound pollution threads in the grape growing process. It collects data on the residual amount of perfluorinated compounds in the soil, the concentration of perfluorinated compounds in irrigation water, and the fluorinated adjuvant components of agricultural inputs as pollution thread factors in the growing process. The second traceability unit is used to trace the perfluorinated compound contamination threads in the static process of wine production. It collects information on the types of fluorinated contact materials of production equipment, the fluorinated coating information of storage containers, and the detection of perfluorinated compounds in processing auxiliary materials, as factors of the contamination threads in the static process of production. The third traceability unit is used to trace the perfluorinated compound contamination threads in the dynamic process of wine production. It collects data on the contact time of fluorinated equipment during material transfer, the amount of perfluorinated compound released from equipment during ultrasonic processes, and the perfluorinated compound settling rate in the production environment as factors of the contamination threads in the dynamic process of production.

[0058] In this embodiment, for example, the collection of perfluorinated compound (PFOC) residues in the soil was carried out by selecting sampling points in the vineyard using a 5-point sampling method, at a depth of 0-20 cm (the main distribution area of ​​grape roots). 500g of soil sample was collected from each sampling point. After removing impurities such as stones and weeds, the residual amount of PFOCs in the soil was detected using high-performance liquid chromatography-mass spectrometry (HPLC-MS / MS). For example, the detected residual amount of PFOS in the soil was 12 ng / kg, and the residual amount of PFHxS was 8 ng / kg.

[0059] The concentration of perfluorinated compounds in irrigation water was collected by taking a 1L water sample at the outlet of the irrigation water source, filtering it through a 0.22μm filter membrane, and then detecting the concentration of perfluorinated compounds in the irrigation water using solid-phase extraction-mass spectrometry. For example, the concentration of PFOS in the irrigation water was found to be 6ng / L, and the concentration of perfluoropentanoic acid (PFPeA) was found to be 4ng / L.

[0060] To collect samples of fluoride-containing adjuvants used in agricultural inputs, three batches each of pesticides and fertilizers were selected. A 10g sample was taken from each batch, and the fluoride-containing adjuvants were extracted using Soxhlet extraction. The types and contents of the fluoride-containing adjuvants were then identified using infrared spectroscopy and mass spectrometry. For example, one batch of pesticide contained perfluorooctyl sulfonate as a fluoride emulsifier at a content of 15 ng / g; another batch of fertilizer contained perfluorohexyl betaine as a fluoride dispersant at a content of 10 ng / g.

[0061] For example, the pollution factors in the planting process are ultimately integrated as follows: Soil: PFOS 12ng / kg, PFHxS 8ng / kg; Irrigation water: PFOS 6ng / L, PFPeA 4ng / L; Agricultural inputs: perfluorooctyl sulfonate ammonium 15ng / g (pesticide), perfluorohexyl betaine 10ng / g (fertilizer)}.

[0062] In this embodiment, for example, the identification of fluorine-containing contact materials in production equipment involves material testing of components in wine production that come into direct contact with materials, such as crushing equipment, pressing equipment, and fermentation tanks. X-ray fluorescence spectrometry is used to analyze the fluorine content in the materials to determine the type of fluorine-containing material. For example, the sealing ring of the crushing equipment is made of fluororubber (fluorine content 30%), and the filter screen of the press is made of polytetrafluoroethylene (fluorine content 76%).

[0063] Information collection on fluorinated coatings of storage containers involves sampling the coatings on the inner walls of tanks storing grape juice and barrels storing finished wine. Scanning electron microscopy-energy dispersive spectroscopy (EDS) is used to analyze the elemental composition of the coatings to determine the material and fluorine content of the fluorinated coating. For example, the inner wall coating of a certain tank is fluorinated polyurethane with a fluorine content of 25%; the inner wall of a certain wine barrel has no fluorinated coating.

[0064] The detection of perfluorinated compounds in processing auxiliary materials involves selecting two batches each of clarifying agents (such as bentonite) and stabilizers (such as vitamin C) used in production, taking 5g samples from each batch, extracting and purifying them, and then using mass spectrometry to detect the perfluorinated compound content. For example, no perfluorinated compounds were detected in a batch of bentonite; a trace amount of PFHxS was detected in a batch of vitamin C, with a concentration of 2ng / g.

[0065] For example, the final contamination thread factors in the static production process are: {Production equipment: fluororubber sealing rings (containing 30% fluorine), polytetrafluoroethylene filter screens (containing 76% fluorine); Storage containers: fluorinated polyurethane coated storage tanks (containing 25% fluorine), fluorine-free coated wine barrels; Processing auxiliary materials: bentonite (no perfluorinated compounds detected), vitamin C (PFHxS2ng / g)}.

[0066] In this embodiment, for example, the contact time of fluorinated material handling equipment is recorded through the production workshop's monitoring system and operation records. This involves calculating the contact time between grapes and fluorinated material equipment, such as fluororubber conveyor belts, from harvest to pressing, as well as the contact time of grape juice with fluorinated pipelines during transport from the press to the fermentation tank, thus accumulating the total contact time. For example, if the contact time between grapes and the fluororubber conveyor belt is 30 minutes, and the contact time between grape juice and the fluorinated pipeline is 20 minutes, the total contact time is 50 minutes.

[0067] The detection of perfluorinated leaching in ultrasonic equipment involves placing a simulated wine sample free of perfluorinated compounds into an ultrasonic device during the ultrasonic extraction process. The sample is processed according to actual production parameters: ultrasonic power 300W and ultrasonic time 20 minutes. The concentration of perfluorinated compounds in the sample after processing is then measured, which is the perfluorinated leaching amount. For example, the detected leaching concentration of PFOS is 1 ng / L, and the PFHxS concentration is 0.8 ng / L.

[0068] The detection of perfluorinated compound (PFOC) sedimentation rate in the production environment involves setting up three sedimentation sampling points in the production workshop, placing sampling trays coated with adsorbent, and continuously sampling for 24 hours. The total amount of PFOC adsorbed on the sampling trays is then measured, and the sedimentation rate (sedimentation rate per unit area and unit time) is calculated. For example, the average sedimentation rate of the three sampling points is... For example, the pollution thread factors in the dynamic production process are ultimately integrated as follows: {Material transfer: 50 min contact time with fluorinated equipment; Ultrasonic process: PFOS release 1 ng / L, PFHxS release 0.8 ng / L; Production environment: Perfluorinated compound settling rate} }

[0069] The beneficial effects of the above technical solution are: by refining the pollution thread factors in the three key links of grape planting, static production, and dynamic production, the specific sources of perfluorinated compound pollution that may occur in each link are identified, providing accurate and comprehensive basic data for subsequent construction of four-dimensional arrays and analysis of residual amounts, effectively solving the problems of vague pollution source tracing and missing key factors in existing technologies, and improving the pertinence and reliability of pollution analysis.

[0070] This invention provides a device for detecting harmful substances in wine, wherein the array construction module further includes: The simulation unit is used to simulate the initial range of theoretical residual concentrations of perfluorinated compounds migrating from the contact medium to the matrix, based on the surface characteristics of the contact medium in each contamination thread, the carbon chain length of the perfluorinated compounds, and the adsorption coefficient of the matrix for the corresponding type of perfluorinated compounds. The deviation calibration unit is used to call a preset cross-matrix pollution feature matching library of perfluorinated compounds and matrix, match historical pollution data that are consistent with the contact medium type and perfluorinated compound type of the current pollution thread, and perform deviation calibration on the initial range of the theoretical residual concentration. The experimental unit is used to conduct customized small-scale experimental verification experiments on the current contamination thread. Among them, the contact conditions in actual production are used as parameters to conduct simulated contact experiments between the matrix and the contact medium of the corresponding contamination thread, and the actual concentration of perfluorinated compounds in the matrix is ​​detected after the experiment. The interval comparison unit is used to compare the actual concentration obtained from the small-scale test with the calibrated theoretical residual concentration range to obtain a suspected pollution residue set that includes the types of perfluorinated compounds, the corrected residual concentration range, and the pollution contact path. An array construction unit is used to construct a four-dimensional array based on the pollution thread factors of the planting stage, the static production stage, and the dynamic production process, and in combination with the suspected pollution retention set corresponding to each pollution thread, the adsorption-desorption efficiency of perfluorinated compounds between stages, and the transfer parameters, to construct a pollution stage-pollution thread factor-cross-stage dynamic migration parameter-suspected pollution retention set.

[0071] In this embodiment, for example, the parameter determination takes the soil-PFOS contamination thread in the planting process as an example, and determines the surface characteristics of the contact medium, i.e., the soil: surface roughness is 5 μm, specific surface area is... The polarity is weak; the carbon chain length of perfluorinated compounds (PFOS) is 8; through literature review and laboratory testing, the adsorption coefficient of soil for PFOS was determined to be 1000, that is, the ratio of the concentration of PFOS in the soil to the concentration in grapes is 1000.

[0072] Based on the above parameters, an adsorption-desorption kinetic model was used for simulation calculations. Assuming the residual PFOS in the soil is 12 ng / kg, the initial theoretical residual concentration range of PFOS in grapes was calculated to be 10-15 ng / kg based on the adsorption coefficient. Considering the error in the adsorption process, the range was set to ±25% of the theoretically calculated value.

[0073] In this embodiment, the matching library call is to call a cross-matrix pollution feature matching library. This database records historical pollution data of soil-PFOS combinations. For example, in a historical case, when the soil PFOS residue was 11 ng / kg, the actual residue concentration in grapes was in the range of 9-13 ng / kg; in another case, when the soil PFOS residue was 13 ng / kg, the actual residue concentration in grapes was in the range of 11-14 ng / kg.

[0074] Interval calibration compares the soil PFOS residue (12 ng / kg) of the current contaminated thread with historical data. It was found that the residue concentration range of the historical data was slightly lower than the initial simulation range. Therefore, the initial range was calibrated to adjust the theoretical residue concentration range of 10-15 ng / kg to 9-14 ng / kg.

[0075] In this embodiment, the experimental design simulates the contact conditions in actual planting. 500g of soil with the same characteristics as the soil in the actual planting area (PFOS residue 12ng / kg) was taken and planted with 3 grape seedlings of the same growth. Irrigation water, light, temperature and other conditions were the same as in actual planting. Grape samples were harvested after 30 days of cultivation.

[0076] The actual concentration was determined by mass spectrometry after pretreatment of the grape sample. The result was 11 ng / kg.

[0077] In this embodiment, the interval comparison is to compare the actual concentration of 11 ng / kg obtained from the small-scale test with the calibrated theoretical residual concentration range of 9-14 ng / kg. It was found that the actual concentration was within this range, so the calibrated range of 9-14 ng / kg was retained.

[0078] Constructing the retention set involves integrating the types of perfluorinated compounds (PFOS), the corrected residual concentration range (9-14 ng / kg), and the contamination exposure pathway (soil → grapes) to form a potential retention set for this contamination pathway. If the actual concentration exceeds the range, for example, if the actual concentration is 16 ng / kg, the deviation percentage is calculated as (16-14) / (14-9) = 40%. The upper limit of the range is then corrected to 14 × (1 + 40%) = 19.6 ng / kg, resulting in the corrected range of 9-19.6 ng / kg. The potential retention set is then constructed.

[0079] In this embodiment, pollution thread factors from the planting stage, static production stage, and dynamic production process are integrated, along with the suspected pollution retention sets corresponding to each pollution thread, such as the retention set of soil-PFOS in the planting stage and the retention set of equipment-PFHxS in the static production stage. The adsorption-desorption efficiency of perfluorinated compounds between stages is also integrated, such as the adsorption-desorption efficiency of soil-grape (30%) and equipment-grape juice (25%). Transfer parameters are also integrated, such as the transfer amount from soil to grapes (8 ng / kg) and from equipment to grape juice (2 ng / kg). This constructs a four-dimensional array of pollution stages, pollution thread factors, cross-stage dynamic migration parameters, and suspected pollution retention sets. For example, one data point is: [Static production stage; Production equipment (fluororubber sealing ring); Adsorption-desorption efficiency 25%, Transfer amount 2 ng / kg; PFHxS: 3-6 ng / kg (equipment → grape juice)].

[0080] The beneficial effects of the above technical solution are: by combining simulation calculation, historical data calibration, and small-scale experimental verification, the suspected pollution residue set of each pollution thread can be accurately determined, ensuring the reliability of the residue concentration range; a four-dimensional array is constructed based on multi-dimensional parameters to realize the structured integration of pollution information, providing high-quality data support for subsequent analysis of the maximum and minimum possible residue amounts, and effectively solving the problems of large prediction deviation of residue range and disordered data integration in existing technologies.

[0081] This invention provides a device for detecting harmful substances in wine, wherein the mass spectrometer screening module includes: The interval extraction unit is used to extract the corrected residual concentration interval of perfluorinated compounds in the suspected pollution retention set corresponding to each pollution thread factor from the four-dimensional array. The weight acquisition unit is used to obtain the contribution weight value of the corresponding contamination thread factor based on the ratio of the product of the contact time and contact area between the contact medium and the matrix of the corresponding contamination thread factor to the sum of the products of the contact media of all contamination thread factors, and in combination with cross-stage dynamic parameters. The multiplication unit is used to multiply the upper and lower limits of the corrected residual concentration range corresponding to the pollution thread factor by the corresponding contribution weight value to obtain the weighted residual concentration range of the corresponding pollution thread factor. The upper limits of the weighted intervals of all factors are summed to obtain the maximum possible residual, and the lower limits are summed to obtain the minimum possible residual. The residual adjustment unit is used to call the actual concentration reference value of the small-scale test. If the deviation between the possible residual value and the actual concentration reference value exceeds the threshold, the response residual will be adjusted.

[0082] In this embodiment, the corrected residual concentration range of perfluorinated compounds in the suspected pollution residue set corresponding to each pollution thread factor is extracted from the four-dimensional array. Assuming there are four key pollution thread factors, their corrected residual concentration ranges are as follows: Planting stage - Soil PFOS: 9-14 ng / kg; Planting stage - Agricultural input PFHxS: 3-8 ng / kg; Static production process - Equipment PFOS: 2-5 ng / kg; Production dynamics - Ultrasonic process PFHxS: 1-3 ng / kg; In this embodiment, the contact intensity of each factor is calculated as follows: Contact intensity = Contact duration × Contact area; Weight value = (Contact intensity of this factor / Sum of contact intensities) × Cross-stage dynamic parameter coefficient. The cross-stage dynamic parameter coefficient is set based on the adsorption-desorption efficiency; the higher the adsorption-desorption efficiency, the larger the coefficient. Here, the coefficients are set to 1.0 for factor 1, 0.9 for factor 2, 0.8 for factor 3, and 0.7 for factor 4. It should be noted that the sum of the weight values ​​is close to 1 due to slight rounding errors during the calculation process.

[0083] In this embodiment, the weighted residual concentration range is calculated by multiplying the upper and lower limits of the corrected residual concentration range of each factor by the corresponding weight value, and then summing the upper limits of all factors and summing the lower limits of all factors to obtain the maximum possible residual and the minimum possible residual.

[0084] In this embodiment, the deviation adjustment is to retrieve the actual concentration reference values of perfluorinated compounds in the small-scale verification experiment, such as the actual concentration of PFOS being 11 ng / kg, the actual concentration of PFHxS being 0.18 ng / kg, and the total actual concentration being 11.18 ng / kg. Calculate the deviation between the maximum possible residue and the actual concentration reference value: (13.69 - 11.18) / 11.18 ≈ 22.5%. If the deviation threshold is set at 10%, then the deviation exceeds the threshold and the response residue needs to be adjusted. For example, reduce the weight value of factor 1 to 0.9, and recalculate the maximum possible residue = 14 × 0.9 + 0.26 + 0.02 + 0.0006 ≈ 12.88 ng / kg, the deviation (12.88 - 11.18) / 11.18 ≈ 15.2%, still exceeding the threshold. Continue to adjust the weight of factor 1 to 0.85, the maximum possible residue = 14 × 0.85 + 0.26 + 0.02 + 0.0006 ≈ 12.18 ng / kg, the deviation (12.18 - 11.18) / 11.18 ≈ 9.0%, meeting the deviation threshold requirement. Finally, determine the maximum possible residue as 12.18 ng / kg and the minimum possible residue as 8.73 ng / kg.

[0085] The beneficial effects of the above technical solution are as follows: By introducing the contribution weight value, comprehensively considering the contact intensity of each pollution thread factor and the influence of cross-link dynamic parameters on the residue amount, the calculation of the maximum and minimum possible residue amounts is more in line with the actual situation; combining the actual concentration reference values of the small-scale verification experiment for deviation adjustment further improves the accuracy of residue amount prediction, effectively solves the problems of one-sided and large deviation in residue amount estimation in the prior art, and provides a reliable basis for subsequent pretreatment process adjustment and detection instrument screening.

[0086] The present invention provides a detection device for harmful substances in wine. The measurement module includes: when the maximum possible residue concentration is u1, the dosage of the extractant is 2 times the volume of the wine sample; when the maximum possible residue concentration < u1, the dosage of the extractant is increased to 3 times the volume of the wine sample, where u1 is the concentration threshold.

[0087] In this embodiment, u1 is the core threshold for guiding the dosage of the extractant, which is set based on the complexity of the wine matrix and the type of perfluorinated compounds. The basic logic is: u1 = (instrument detection limit × 10). Ensure that the dosage of the extractant is increased at low residues to improve the enrichment efficiency. Taking the detection limit of 0.05 ng / kg of the triple quadrupole tandem mass spectrometer as the reference instrument, the reference u1 = 0.05 ng / kg × 10 = 0.5 ng / kg.

[0088] Regarding the differences in the sugar and organic acid contents of different matrices that will affect the interaction efficiency between the extractant and the target compound, the adjustment coefficients are shown in Table 5: Table 5 Adjustment comparison table of matrix types

[0089] In this embodiment, the differences in carbon chain length and polarity of different perfluorinated compounds affect their residual stability in the matrix, and the adjustment coefficients are shown in Table 6: Table 6. Adjustment Comparison Table for Perfluorinated Compound Types

[0090] In this embodiment, the overall u1 = baseline u1 × matrix adjustment factor × compound type adjustment factor: Example 1: For the detection of PFBA (short chain, weakly polar) in sweet wine, the baseline u1 = 0.5 ng / kg, matrix adjustment factor = 1.2, compound type adjustment factor = 1.2, and the overall u1 = 0.5 × 1.2 × 1.2 = 0.72 ng / kg. → If the maximum possible residual concentration is ≥ 0.72 ng / kg, the amount of extractant used = sample volume × 2; if < 0.72 ng / kg, the amount of extractant used = sample volume × 3.

[0091] Example 2: For detecting PFHxA (medium chain, medium polarity) in dry red wine, the baseline u1 = 0.5 ng / kg, matrix adjustment factor = 1.1, compound type adjustment factor = 1.1, and the overall u1 = 0.5 × 1.1 × 1.1 = 0.605 ng / kg. → Adjust the amount of extractant according to this threshold to ensure that the medium chain compound is fully extracted in the high organic acid matrix.

[0092] The beneficial effects of the above technical solution are as follows: the amount of extractant used in the pretreatment process is dynamically adjusted according to the maximum possible residual concentration of perfluorinated compounds, thereby achieving targeted optimization of the extraction process: when the residual concentration is high, the conventional dosage is used to ensure the extraction effect while avoiding reagent waste; when the residual concentration is low, the dosage is increased to improve the extraction efficiency and ensure that the target compound is fully enriched. This solves the problem of fixed extractant dosage in existing technologies, which cannot be adapted to samples with different residual levels, and improves the flexibility and effectiveness of the pretreatment process, thus ensuring the accuracy of subsequent detection results.

[0093] This invention provides a device for detecting harmful substances in wine, wherein the result comprehensive analysis module includes: The vector construction unit is used to extract features from the measurement matrix according to each standard in the measurement requirement standard to obtain the first feature and construct the feature vector. At the same time, it performs external interference analysis on the measurement results of the blank experimental group to obtain the interference vector. A limit processing unit is used to perform limit processing on the feature vector to obtain a limit processing vector. The vector optimization unit is used to perform position difference analysis on each element in the limit processing vector according to each interference factor in the interference vector, determine the posterior distribution of each element, and correct the corresponding elements according to the posterior distribution to obtain the optimized vector, which is regarded as a single detection result.

[0094] In this embodiment, the i-th element in the limit processing vector is considered. Combined with the kth interference factor in the interference vector Based on the positional difference analysis results, a Bayesian model is used to construct the i-th element. posterior distribution ,in, Given a prior distribution, vector elements are treated using the limit method. The mean is expressed as the standard deviation of the measurement of the element. The variance follows a normal distribution, i.e. ; The interference factors are obtained based on the likelihood distribution and the potential difference analysis. Compared with the true value The relationship between them, For position difference, construct with The mean is expressed as the standard deviation of the interference measurement. It follows a normal distribution with variance, and The influence coefficient of the kth interfering lease is determined by the interference intensity of the blank experimental group; This is the normalization constant; In this embodiment, the posterior distribution The expected value is used as an estimate of the true value, and the i-th element of the limit processing vector is corrected to obtain the optimized vector. The i-th element : .

[0095] In this embodiment, the feature vector construction uses the measurement matrix of a mass spectrometer as an example. Feature information is extracted according to the measurement requirements, including the PFOS concentration (7.2 ng / kg, 6.8 ng / kg, 7.0 ng / kg), PFHxS concentration (4.5 ng / kg, 4.3 ng / kg, 4.6 ng / kg), and specific response intensity (P0) of three measurements. ), construct the feature vector V: .

[0096] Interference vector construction involved analyzing the measurement results of the blank experimental group and identifying three main interference factors: the interference intensity of trace PFOS in the environment (0.1 ng / kg), the interference intensity of instrument noise, and so on. The interference intensity of trace PFHxS in the reagent is 0.08 ng / kg. An interference vector D is constructed as follows: D = [0.1 (PFOS environmental interference), 0.08 (PFHxS reagent interference)]. (Instrument noise interference).

[0097] In this embodiment, the extreme processing feature vector is the element in the feature vector V subjected to extreme processing. For example, the extreme range of PFOS concentration is 6.8-7.2 ng / kg, and the median value of 7.0 ng / kg is taken as the representative value of this group of elements; the median value of PFHxS concentration is taken as 4.47 ng / kg; and the median value of the specific response intensity is taken as... The limit processing vector is obtained. =[7.0,4.47, The values ​​are arranged in the order of PFOS concentration, PFHxS concentration, and specific response intensity.

[0098] In this embodiment, position difference analysis is the analysis of the interference vector D and the limit processing vector. The positional difference relationship is used to determine the corresponding interference factors in D. The influence of each element: D1 (PFOS environmental interference) corresponds to (PFOS concentration), position difference is 0.1 ng / kg; D2 (PFHxS reagent interference) corresponds to (PFHxS concentration), position difference is 0.08 ng / kg; D3 (instrument noise interference) corresponds to (Specific response strength), position difference is .

[0099] The posterior distribution is constructed using a Bayesian model. The posterior distribution of each element in the set.

[0100] In this embodiment, the expectation of the posterior distribution is used as an estimate of the true value of the element to correct each element in the limit processing vector.

[0101] The beneficial effects of the above technical solution are as follows: by constructing core data vectors through feature extraction and limit processing, identifying interference factors and performing position difference analysis in conjunction with blank experiments, and using Bayesian models to correct interference effects, the influence of external interference on detection results is effectively eliminated, improving the accuracy and reliability of single detection results; this data processing method is logically rigorous and highly operable, solving the problems of difficulty in quantifying interference factors and easy distortion of detection results in existing technologies, laying a solid foundation for the comprehensive analysis of subsequent multi-instrument detection results.

[0102] This invention provides a device for detecting harmful substances in wine, wherein the result comprehensive analysis module further includes: The coefficient determination unit is used to determine the detection fit coefficient of the perfluorinated compound corresponding to each single detection result based on the carbon chain length and polarity of the target perfluorinated compound in the four-dimensional array. At the same time, it determines the measurement condition fit coefficient based on the chromatographic resolution and mass spectrometry response intensity in the measurement matrix corresponding to each single detection result. The weight calculation unit is used to calculate the confidence weight of a single detection result based on the product of the detection fit coefficient and the measurement condition fit coefficient. The elimination unit is used to calculate the mean and standard deviation of each perfluorinated compound in all single test results. If the absolute value of the difference between the concentration value of any perfluorinated compound and the corresponding mean in the corresponding single test result is greater than twice the standard deviation, and the corresponding confidence weight is less than the preset weight, then the concentration value of the corresponding perfluorinated compound in the corresponding single test result is eliminated. The final calibration unit is used to perform weighted calculations on the concentration values ​​of each retained perfluorinated compound to obtain the initial test results. At the same time, it retrieves the actual concentration reference values ​​of the corresponding perfluorinated compounds in the small-scale experimental verification to calibrate the initial test results, obtain the final test results, and output pollution control suggestions.

[0103] In this embodiment, taking the target perfluorinated compounds as PFOS (carbon chain 8, highly polar) and PFHxS (carbon chain 6, moderately polar), and a single detection result from mass spectrometers A and B as an example: Retrieve the properties of PFOS and PFHxS from the four-dimensional array: PFOS (carbon chain 8, strong polarity) and PFHxS (carbon chain 6, medium polarity).

[0104] In this embodiment, the fit coefficient = carbon chain length fit score × polarity fit score, and the core parameter classification and scoring criteria are shown in Table 7: Table 7 Core Parameter Classification and Scoring Criteria

[0105] In this embodiment, the fit coefficient = (chromatographic resolution score × 50%) + (mass spectrometry response intensity score × 50%). It should be noted that the weighting is based on the core logic that resolution determines qualitative accuracy and response intensity determines quantitative accuracy. This weighting is adjusted according to the detection priority, and the grading and scoring criteria of the core parameters are shown in Table 8. Table 8. Core Parameter Classification and Scoring Criteria

[0106] In this embodiment, the confidence weight = detection fit coefficient × measurement condition fit coefficient.

[0107] In this embodiment, the mean and standard deviation are calculated: PFOS concentration: Mass spectrometer A: 7.0 ng / kg, B: 7.1 ng / kg, mean (7.0+7.1) / 2=7.05 ng / kg, standard deviation ≈0.035 ng / kg; PFHxS concentration: Mass spectrometer A: 4.5 ng / kg, B: 4.6 ng / kg, mean (4.5+4.6) / 2=4.55 ng / kg, standard deviation ≈0.035 ng / kg.

[0108] Set the preset weight to 0.85 to remove outliers: PFOS: If the difference between the two concentration values ​​and the mean is less than 2 × 0.035 and the confidence weight is greater than 0.85, both values ​​are retained. PFHxS: The confidence weight of mass spectrometer B is 0.8372 < 0.85, so its concentration value of 4.6 ng / kg is discarded.

[0109] In this embodiment, the initial weighted calculation result is as follows: PFOS: (7.0×0.912+7.1×0.893) / (0.912+0.893)≈7.05ng / kg; PFHxS: Only 4.5 ng / kg of mass spectrometer A is retained.

[0110] Reference values ​​obtained from small-scale tests: PFOS: 7.0 ng / kg, PFHxS: 4.4 ng / kg. Final calibration results: PFOS: 7.05×(7.0 / 7.05)=7.0ng / kg; PFHxS: 4.5×(4.4 / 4.5)=4.4ng / kg.

[0111] The proposed corrective measures are as follows: PFOS originates from residues in the planting soil, and it is recommended to purify the planting soil; PFHxS originates from fluoride-containing agricultural adjuvants, and it is recommended to replace them with fluoride-free adjuvants.

[0112] The beneficial effects of the above technical solution are: by quantifying the adaptability and condition fit of the detection to determine the reliability of the results, eliminating low-weight outliers and combining small-scale experimental calibration, the accuracy and stability of the detection results are greatly improved; the final rectification suggestions accurately point to the source of pollution, providing scientific and practical technical guidance for the control of perfluorinated compound pollution in wine production.

[0113] This invention provides a method for detecting harmful substances in wine, such as... Figure 2 As shown, it includes: Step 1: Trace the perfluorinated compound (PFCC) contamination threads of grapes in the planting, static production, and dynamic production processes, and collect the adsorption-desorption efficiency and transfer parameters of PFCCs in each process. Combine this with the suspected contamination retention set of each contamination thread to construct a four-dimensional array. Step 2: Based on the analysis of the maximum and minimum possible residues of perfluorinated compounds in wine using the four-dimensional array, call the perfluorinated compound carbon chain length-mass spectrometry response specificity database, lock the determination requirements standard, and select the mass spectrometer to be measured from the standard-measurement comparison table that meets the determination requirements standard. The number of the mass spectrometer to be measured is at least 2, and the determination requirements standard includes: detection limit and specific response intensity. Step 3: Pre-treat the wine and divide the pre-treated wine into n1×n2 portions as measurement samples. Perform n2 measurements on each sample based on the mass spectrometer to be measured, and construct a measurement matrix for each mass spectrometer to be measured, where n1 is the number of mass spectrometers to be measured. The measurement matrix includes: the types and contents of perfluorinated compounds in each measurement, and is associated with the determination conditions for each measurement. The determination conditions include: chromatographic conditions and mass spectrometry conditions. Step 4: Analyze the corresponding measurement matrix based on the aforementioned measurement requirements standards, and obtain a single detection result by combining the measurement results of the blank experimental group of the corresponding mass spectrometer to be measured. Then, conduct a comprehensive analysis of all single detection results and output the final detection result and suggestions for rectifying the pollution process.

[0114] The beneficial effects of the above technical solution are as follows: a four-dimensional array is constructed through full-process pollution source tracing to achieve comprehensive integration of pollution information; a mass spectrometer is selected and adapted based on compound characteristics and residual amounts to ensure the targeting of detection; a measurement matrix is ​​constructed through multiple measurements by multiple instruments, and interference is eliminated by combining blank experiments to improve the accuracy and reliability of detection results; the final output of detection results and suggestions for rectification of pollution links provides precise guidance for quality control in the wine production process and effectively reduces the risk of perfluorinated compound residues.

[0115] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A device for detecting harmful substances in wine, characterized in that, The method comprises the following steps: An array construction module is used to trace the perfluorinated compound pollution threads of grapes in the planting link, the production static link and the production dynamic process respectively, and collect the adsorption-desorption efficiency and transfer amount parameters of perfluorinated compounds between each link, and combine the pollution suspected retention set of each pollution thread to construct a four-dimensional array; A mass spectrometer screening module is used to analyze the maximum possible residue and the minimum possible residue of perfluorinated compounds in wine based on the four-dimensional array, call a perfluorinated compound carbon chain length-mass spectrum response specificity database, lock a determination requirement standard, and screen a to-be-measured mass spectrometer that meets the determination requirement standard from a standard-measurement control table, wherein the number of the to-be-measured mass spectrometer is at least two, and the determination requirement standard comprises a detection limit and a specific response intensity; A measurement module is used to pretreat the wine, divide the pretreated wine into n1×n2 parts as measurement samples, and perform n2 measurements on each sample based on each to-be-measured mass spectrometer to construct a measurement matrix for each to-be-measured mass spectrometer, wherein n1 is the number of to-be-measured mass spectrometers, and the measurement matrix comprises the type and component content of perfluorinated compounds in each measurement and is associated with the determination conditions of each measurement, and the determination conditions comprise chromatographic conditions and mass spectrometric conditions; A result comprehensive analysis module is used to analyze the corresponding measurement matrix based on the determination requirement standard, obtain a single detection result in combination with the measurement result of a blank experiment group of the corresponding to-be-measured mass spectrometer, and comprehensively analyze all single detection results and output a final detection result and a pollution link improvement suggestion.

2. The wine hazardous substance detection device according to claim 1, characterized by, The array construction module comprises: A first tracing unit is used to trace the perfluorinated compound pollution thread of the grape planting link, and collect the soil perfluorinated compound residue, the irrigation water perfluorinated compound concentration and the fluorine-containing additive component of the planted agricultural means as the planting link pollution thread factors; A second tracing unit is used to trace the perfluorinated compound pollution thread of the wine production static link, and collect the fluorine-containing contact material type of the production equipment, the fluorine-containing coating information of the storage container and the perfluorinated compound detection situation of the processing auxiliary material as the production static link pollution thread factors; A third tracing unit is used to trace the perfluorinated compound pollution thread of the wine production dynamic process, and collect the fluorine-containing equipment contact time of material transfer, the perfluorinated compound emission amount of the equipment in the ultrasonic process and the perfluorinated compound settling rate of the production environment as the production dynamic process pollution thread factors.

3. The wine hazardous substance detection device according to claim 2, characterized by, The array construction module further comprises: An analog unit is used to simulate the initial interval of the theoretical residue concentration of perfluorinated compounds migrated from the contact medium to the substrate based on the surface characteristics of the contact medium, the carbon chain length of perfluorinated compounds and the adsorption coefficient of the corresponding perfluorinated compound to the substrate; A deviation calibration unit is used to call a preset perfluorinated compound-substrate cross-substrate pollution characteristic matching library, match the historical pollution data consistent with the contact medium type and the perfluorinated compound type of the current pollution thread, and calibrate the deviation of the initial interval of the theoretical residue concentration. An experimental unit is configured to carry out customized small-scale test experiments on current pollution threads, in which the actual production contact conditions are used as parameters to carry out simulation contact experiments of substrates and corresponding pollution thread contact media, and the actual concentration of perfluorinated compounds in the substrate after the experiment is detected; An interval comparison unit is configured to compare the actual concentration obtained by the small-scale test with the calibrated theoretical residual concentration interval to obtain a pollution suspected residual set containing perfluorinated compound types, a modified residual concentration interval, and a pollution contact path; An array construction unit is configured to construct a four-dimensional array of pollution link-pollution thread factor-cross-link dynamic migration parameter-pollution suspected residual set based on pollution thread factors of the planting link, production static link, and production dynamic process, and in combination with the pollution suspected residual set corresponding to each pollution thread, the adsorption-desorption efficiency of perfluorinated compounds between links, and the transfer amount parameters.

4. The wine hazardous substance detection device according to claim 3, characterized by, The mass spectrometer screening module comprises: An interval extraction unit is configured to extract the modified residual concentration interval of perfluorinated compounds in the pollution suspected residual set corresponding to each pollution thread factor from the four-dimensional array; A weight acquisition unit is configured to obtain the contribution weight value of the corresponding pollution thread factor based on the product of the contact time and contact area of the contact medium and the substrate corresponding to the pollution thread factor, and in combination with the cross-link dynamic parameters; A multiplication unit is configured to multiply the upper limit and lower limit of the modified residual concentration interval corresponding to the corresponding pollution thread factor by the corresponding contribution weight value to obtain the weighted residual concentration interval of the corresponding pollution thread factor, and sum the upper limits of the weighted intervals of all factors to obtain the maximum possible residual, and sum the lower limits to obtain the minimum possible residual; A residual adjustment unit is configured to call the actual concentration reference value of the small-scale test experiment, and adjust the response residual if the deviation of the possible residual from the actual concentration reference value exceeds the threshold.

5. The wine hazardous substance detection device according to claim 1, characterized by, The measuring module comprises: when the maximum possible residual concentration u1, the amount of extractant is 2 times the volume of the wine sample; When the maximum possible residual concentration u1, the amount of extraction agent is increased to 3 times the volume of the wine sample, wherein u1 is the concentration threshold.

6. The wine hazardous substance detection device according to claim 1, characterized by, The result comprehensive analysis module comprises: A vector construction unit is configured to respectively extract features from the measurement matrix according to each standard in the measurement requirement standard to obtain a first feature, and construct a feature vector, and at the same time, analyze the measurement results of the blank experiment group for external interference to obtain an interference vector; A limit processing unit is configured to perform limit processing on the feature vector to obtain a limit processing vector; A vector optimization unit is configured to analyze the bit difference of each element in the limit processing vector according to each interference factor in the interference vector to determine the posterior distribution of each element, and correct the corresponding elements according to the posterior distribution to obtain an optimized vector, which is regarded as a single detection result.

7. The wine spoilage detection device of claim 6, wherein, The result comprehensive analysis module further comprises: A coefficient determination unit is configured to determine the detection adaptation coefficient of each single detection result corresponding to the perfluorinated compound based on the carbon chain length and polarity of the target perfluorinated compound in the four-dimensional array, and at the same time, determine the determination condition fitting coefficient based on the chromatographic separation degree and mass spectrometric response intensity in the measurement matrix corresponding to each single detection result; a weight calculation unit configured to calculate a weight of reliability of each single detection result based on a product of the detection adaptation coefficient and the determination condition fitting coefficient; a rejection unit configured to calculate a mean value and a standard deviation of each perfluorinated compound in all single detection results, and to reject a concentration value of the corresponding perfluorinated compound in the corresponding single detection result if an absolute value of a difference between the concentration value and the corresponding mean value is greater than 2 times the standard deviation and the weight of reliability of the corresponding single detection result is less than a preset weight; a final calibration unit configured to perform weighted calculation on the concentration value of each perfluorinated compound that is retained to obtain an initial detection result, and to calibrate the initial detection result by using an actual concentration reference value of the corresponding perfluorinated compound in the small-scale verification experiment to obtain a final detection result and output a pollution link rectification suggestion.

8. A method for detecting a harmful substance in wine, characterized by, comprising: Step 1: tracing the perfluorinated compound pollution threads of the grape in the planting link, the production static link and the production dynamic process, respectively, collecting the adsorption-desorption efficiency and the transfer amount of the perfluorinated compound between each link, and combining the pollution suspected storage set of each pollution thread to construct a four-dimensional array; Step 2: analyzing the maximum possible residue and the minimum possible residue of the perfluorinated compound in the wine based on the four-dimensional array, calling a carbon chain length-mass spectrum response specificity database of the perfluorinated compound, locking a determination requirement standard, and screening a to-be-measured mass spectrometer that meets the determination requirement standard from a standard-measurement control table, wherein the number of the to-be-measured mass spectrometer is at least 2, and the determination requirement standard includes a detection limit and an exclusive response intensity; Step 3: pretreating the wine, dividing the pretreated wine into n1×n2 parts as measurement samples, and performing n2 measurements on each sample based on each to-be-measured mass spectrometer to construct a measurement matrix for each to-be-measured mass spectrometer, wherein n1 is the number of the to-be-measured mass spectrometer, and the measurement matrix includes the types and component contents of the perfluorinated compound in each measurement and is associated with the determination conditions of each measurement, and the determination conditions include chromatographic conditions and mass spectrometric conditions; Step 4: analyzing the corresponding measurement matrix based on the determination requirement standard, obtaining single detection results in combination with measurement results of a blank experiment group of the corresponding to-be-measured mass spectrometer, and comprehensively analyzing all single detection results to output a final detection result and a pollution link rectification suggestion.

Images