Accurate detection method for content of isomers in LAS

Abstract

The invention discloses a method for accurately detecting the content of isomers in LAS. The method comprises the following steps: step 1, performing multiple pretreatment on isotope coding; step 2, a four-stage chromatographic separation system; step 3, dynamic mass spectrum imaging detection; step 4, deep learning data analysis; 5, accurately calculating the content of the isomer; step 6, quality control and result verification; the method not only solves the problem of spatial isomerism separation, improves the sensitivity of trace detection, but also effectively reduces the redundancy of multi-dimensional data, and significantly improves the accuracy and reliability of the content detection of isomers in LAS. Through multiple pretreatment of isotope coding, the matrix interference of the sample is effectively reduced, and the detection specificity is improved; due to the application of the four-stage chromatographic separation system, effective separation of each isomer is realized; dynamic mass spectrum imaging detection is combined with deep learning data analysis, so that the detection sensitivity and accuracy are further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of detecting the isomer content in LAS, and specifically to a precise detection method for the isomer content in LAS. Background Art

[0002] Isomers in LAS refer to compounds with the same molecular formula but different structures. Isomers are compounds with the same chemical composition but different structures, having the same chemical formula but different arrangements of atoms in space, which leads to significant differences in their physical and chemical properties.

[0003] The existing patent publication number CN112557572A discloses a method for detecting the isomer content in Cbz-Glu, belonging to the field of chemical analysis. Specifically, the high-performance liquid chromatography conditions are as follows: the chromatographic column is CHIRALCEL OX-3, the detection wavelength is 210 nm, the column temperature is 35 °C, the mobile phase is a mixture of n-hexane and absolute ethanol, the flow rate of the mobile phase is 0.6 ml / min, and the elution time is 40 min. It can effectively separate Cbz-L-Glu from its isomer Cbz-D-Glu, thereby detecting the content of the isomer Cbz-D-Glu in terms of time;

[0004] The existing patent publication number CN105424828A discloses a method for detecting the enantiomeric isomer content in levocarnitine. Using a series of benzoyl halide compounds as derivatizing reagents to derivatize levocarnitine, and then detecting by HPLC, and calculating the content of enantiomers by area normalization method. Detecting the enantiomeric isomer content in levocarnitine has high repeatability, precision and accuracy, which is of great significance for controlling the quality of levocarnitine raw materials and ensuring drug safety. In addition, using a series of benzoyl halide compounds as derivatizing reagents has a low cost. TLC method monitoring the derivatization process shows that the derivatization reaction is complete, and levocarnitine and its enantiomeric isomers are all converted into corresponding derivatives, thus ensuring the accuracy of the detection results.

[0005] However, the above-mentioned existing patents all have the following defects when detecting the isomers in LAS:

[0006] (1) Problem of spatial isomer separation: The hydrophobic constant difference between the 2-position and 3-position isomers of alkylbenzene sulfonate is only 0.07 log P, and the resolution (R = 1.3) of the traditional C18 column cannot meet the requirement of baseline separation;

[0007] (2) Trace detection limitation: The concentration of LAS isomers in environmental samples is often lower than 50 μg / L, and the signal-to-noise ratio of the existing MS / MS detection after matrix interference (S / N < 10) is difficult to accurately quantify;

[0008] (3) Multidimensional data redundancy: Traditional two-dimensional chromatography-MS data has insufficient peak capacity (<5000), resulting in an overlap rate of signals of isomers with more than 12 carbon chains exceeding 30%. Summary of the Invention

[0009] The purpose of the present invention is to provide a precise detection method for the content of isomers in LAS to solve the problems raised in the above background technology.

[0010] To achieve the above purpose, the present invention provides the following technical solution: A precise detection method for the content of isomers in LAS, comprising the following steps:

[0011] Step 1. Isotope-coded multiple pretreatment: Specifically divided into seven sub-steps, namely stable isotope labeling, dual-column purification, microwave-assisted derivatization, freeze centrifugation stratification, nanomaterial enrichment, on-line solid-phase microextraction, and membrane injection pretreatment;

[0012] Step 2. Four-stage chromatography separation system: A four-stage separation system including hydrophilic interaction-chiral exclusion-capillary electrophoresis-nano-LC;

[0013] Step 3. Dynamic mass spectrometry imaging detection: Adopt a mass spectrometry imaging detection technology with dynamic switching of dual ion sources;

[0014] Step 4. Deep learning data analysis: An overlapping peak analysis algorithm based on a three-dimensional U-Net network;

[0015] Step 5. Precise calculation of isomer content: Input the results obtained from deep learning data analysis into a pre-constructed isomer content calculation model. This model trains and optimizes the analysis data through machine learning algorithms, and finally outputs the precise content of isomers in LAS;

[0016] Step 6. Quality control and result verification: To ensure the accuracy and reliability of the detection results, a series of quality control measures are required, including calibration with standard products, setting parallel samples for repeatability verification, and cross-verification using third-party methods; at the same time, analyze and correct the errors that may be introduced during the detection process to ensure that the isomer content data in the final report is accurate and error-free.

[0017] Preferably, the stable isotope labeling in Step 1 is specifically: Add 10 μL 13 C6]-2-C12-LAS internal standard, 100 μg / mL, vortex for 30 s;

[0018] The dual-column purification is specifically: Pass through a Florisil column and a SAX column in sequence. The Florisil column is used to remove lipophilic impurities, and the SAX column is used to adsorb anionic interfering substances;

[0019] Microwave-assisted derivatization is specifically as follows: Add N,O-bis(trimethylsilyl)acetamide (BTSA) and perform microwave treatment at 120 °C for 5 min; the power of the microwave-assisted derivatization step is 300 - 600 W, the time is 4 - 6 min, and the volume ratio of the derivatization reagent to the sample is 1:20 - 1:50;

[0020] Freezing and centrifuging for layering is specifically as follows: Freeze at -40 °C for 30 min, centrifuge at 15000 rpm for 20 min, and collect the organic phase;

[0021] Nanomaterial enrichment is specifically as follows: Add 10 mg of MOF-74-NH2 material, shake for adsorption for 2 h, and elute with methanol 3 times;

[0022] On-line solid-phase microextraction is specifically as follows: The SPME fiber, the solid-phase coating material is PDMS / DVB, the coating thickness is 50 / 30 μm, and headspace extraction is performed at 70 °C for 15 min;

[0023] Membrane injection pretreatment is specifically as follows: Filter through a 0.05 μm alumina membrane to remove derivatization by-products.

[0024] Preferably, the four-stage chromatographic separation system in the second step is specifically as follows:

[0025] The first dimension is hydrophilic interaction chromatography:

[0026] The chromatographic column is Shodex Asahipak HILIC-N, the length and inner diameter of the chromatographic column are 300×4.6 mm, and the inner diameter of the chromatographic column is 5 μm;

[0027] Mobile phase gradient: Acetonitrile 95% → 60% + 10 mM ammonium acetate pH 4.2, and carbon chain grouping is completed in 80 min;

[0028] The second dimension is chiral exclusion chromatography:

[0029] The chromatographic column is Chiralcel OD-RH, the length and inner diameter of the chromatographic column are 250×2.0 mm, and the inner diameter of the chromatographic column is 3 μm;

[0030] Mobile phase: Ethanol - water, the proportion of ethanol - water is 90:10, containing 0.1% diethylamine, and isocratic elution is used to separate positional isomers;

[0031] The third dimension is capillary electrophoresis:

[0032] The inner diameter and effective length of the capillary are 50 μmi.d.×50 cm, and the coating is polyethylene glycol;

[0033] The separation voltage is 25 kV, the temperature is 20 °C, and separation of the electrophoretic mobility differences of isomers is achieved;

[0034] The fourth dimension is nano-LC:

[0035] The chromatographic column is Acquity UPLC BEH C18, with a length and inner diameter of 100×0.15 mm and an inner diameter of the column packing of 1.7 μm;

[0036] Gradient: At 0 minutes when injection starts, in the mobile phase, the proportion of methanol is 10%, containing 0.1% formic acid. As time goes by, the proportion of methanol continuously increases and reaches 90% at 15 minutes.

[0037] Preferably, the dynamic mass spectrometry imaging detection in step three specifically includes ion source configuration and mass analyzer mode:

[0038] Ion source configuration: Atmospheric pressure chemical ionization + electrospray dual-source switching;

[0039] Atmospheric pressure chemical ionization ionizes solvent molecules in the mobile phase through corona discharge, and then collides with sample molecules to generate ions;

[0040] Electrospray dual-source switching atomizes liquid samples into charged droplets through a high-voltage electric field and generates ions after evaporation;

[0041] Dual-source switching 5 times per second: Switching 5 times per second rapidly alternates the acquisition of data from the two ion sources in a single experiment, avoiding manual replacement of hardware;

[0042] Imaging parameters are spatial resolution of 50 μm: The instrument can distinguish two adjacent points on the sample surface 50 μm apart, ensuring the details of the molecular distribution image; pixel point cumulative time of 100 ms: The signal acquisition time for each pixel point is 0.1 second;

[0043] Mass analyzer mode: Orbitrap resolution of 120,000 (m / z 200), mass accuracy error < 1 ppm; linear ion trap performs MS 3 fragmentation to obtain the fingerprint spectrum of the third-level fragment ions, multi-stage mass spectrometry MS 3 .

[0044] High resolution: A resolution of 120,000 (at m / z 200) means that the instrument can accurately distinguish ions with extremely small differences in mass-to-charge ratio (such as m / z 200.0001 and 200.0002), reducing isotope or background interference.

[0045] Mass accuracy: Error < 1 ppm, which can accurately deduce the molecular formula (for example, the allowable error range for m / z 200 is ±0.0002 Da).

[0046] Preferably, the overlapping peak resolution algorithm based on the three-dimensional U-Net network specifically includes a three-dimensional peak detection algorithm, a background correction model, and an isomer tracing algorithm; among them,

[0047] The three-dimensional peak detection algorithm is based on the U-Net neural network to identify overlapping peaks in the three-dimensional space of chromatography-electrophoresis-mass spectrometry. The training dataset includes the multidimensional chromatographic behaviors of more than 1000 LAS isomer standards;

[0048] The input features of the background correction model are retention time, mass-to-charge ratio, isotope peak intensity ratio, and matrix fingerprint; the output is the corrected absolute quantitative concentration, R 2 ≥0.995, n = 5000 test set;

[0049] The formula for the isomer traceability algorithm is:

[0050] Among them, I test is the sample ion intensity, and S ref is the signal of the standard product spectral library.

[0051] Preferably, in the four-stage chromatographic separation system:

[0052] The coating material of the third-dimensional capillary electrophoresis is zwitterionic polymer, and the zeta potential ≤ -20 mV;

[0053] The column oven of the fourth-dimensional nano-flow LC is equipped with a gradient cooling device, and the temperature fluctuation ≤ ±0.1 °C.

[0054] Preferably, the dynamic switching period of the dual ion source ≤ 200 ms, and ≥ 50 mass spectra are collected for each pixel.

[0055] Preferably, the accurate calculation of the isomer content in step five is specifically to input the overlapping peak analysis results obtained by parsing the deep learning data, the isomer peak areas separated by each-dimensional chromatography, and the standard curve data into a pre-constructed isomer content calculation model. The model uses machine learning algorithms such as multiple linear regression or support vector machine to train and optimize the input data to establish a quantitative relationship between the isomer peak area and the content; through model calculation, the accurate content of each isomer in LAS is finally output, including relative content and absolute content, providing accurate and reliable data support for subsequent isomer analysis.

[0056] Preferably, the quality control and result verification in step six are specifically as follows: quantitative analysis is carried out using the internal standard method to ensure the accuracy of data; high-purity standards are used for calibration to reduce errors; blank control experiments are set up to monitor potential contamination during the experiment; at the same time, statistical analysis of the experimental data is carried out, including calculating the mean, standard deviation and coefficient of variation to evaluate the stability and reliability of the data; in addition, regular participation in external quality assessment programs and comparison with peer laboratories are also carried out to continuously improve the detection level; the internal standard method specifically refers to adding an internal standard with a known concentration to the sample, performing pretreatment and detection together with the component to be measured, and calculating the content of the component to be measured by comparing the peak area or peak height ratio of the component to be measured with that of the internal standard.

[0057] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0058] This method not only solves the problem of spatial isomer separation, improves the sensitivity of trace detection, but also effectively reduces the redundancy of multidimensional data, and significantly improves the accuracy and reliability of the detection of isomer content in LAS. Through isotope-coded multiple pretreatment, the interference of the sample matrix is effectively reduced, and the specificity of detection is improved; the application of the four-stage chromatographic separation system realizes the effective separation of various isomers; the combination of dynamic mass spectrometry imaging detection and deep learning data analysis further improves the sensitivity and accuracy of detection; and the strict implementation of quality control and result verification measures ensures the reliability and traceability of the final detection results. Therefore, the present invention provides a new solution for the accurate detection of isomer content in LAS, and has broad application prospects and market value. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is the flow chart of the preparation method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0061] Please refer to Figure 1 , the present invention provides a method for accurately detecting the isomer content in LAS, including the following steps:

[0062] Step 1. Isotope-coded multiple pretreatment: It is specifically divided into seven sub-steps, namely stable isotope labeling, double-column purification, microwave-assisted derivatization, freezing and centrifugal stratification, nanomaterial enrichment, on-line solid-phase microextraction and membrane injection pretreatment; the stable isotope labeling is specifically as follows: adding 10 μL13 Internal standard of C6]-2-C12-LAS, 100 μg / mL, vortex for 30 s;

[0063] The double-column purification is specifically as follows: sequentially pass through a Florisil column and a SAX column. The Florisil column is used to remove fat-soluble impurities, and the SAX column is used to adsorb anionic interfering substances;

[0064] The microwave-assisted derivatization is specifically as follows: add N,O-bis(trimethylsilyl)acetamide (BTSA), and perform microwave treatment at 120 °C for 5 min; the power of the microwave-assisted derivatization step is 300 - 600 W, the time is 4 - 6 min, and the volume ratio of the derivatization reagent to the sample is 1:20 - 1:50;

[0065] The freezing and centrifugal stratification is specifically as follows: freeze at -40 °C for 30 min, centrifuge at 15000 rpm for 20 min, and collect the organic phase;

[0066] The nanomaterial enrichment is specifically as follows: add 10 mg of MOF-74-NH2 material, shake and adsorb for 2 h, and elute with methanol 3 times;

[0067] The on-line solid-phase microextraction is specifically as follows: an SPME fiber, the solid-phase coating material is PDMS / DVB, the coating thickness is 50 / 30 μm, and perform headspace extraction at 70 °C for 15 min;

[0068] The membrane injection pretreatment is specifically as follows: filter through a 0.05 μm alumina membrane to remove derivatization by-products;

[0069] Step 2: Four-stage chromatographic separation system: a four-stage separation system including hydrophilic interaction-chiral exclusion-capillary electrophoresis-nano-LC; in the four-stage chromatographic separation system: the coating material of the third-dimensional capillary electrophoresis is an amphoteric ion polymer, and the zeta potential ≤ -20 mV; the column oven of the fourth-dimensional nano-LC is equipped with a gradient cooling device, and the temperature fluctuation ≤ ±0.1 °C; the four-stage chromatographic separation system is specifically as follows:

[0070] The first dimension is hydrophilic interaction chromatography:

[0071] The chromatographic column is Shodex Asahipak HILIC-N, the length and inner diameter of the chromatographic column are 300×4.6 mm, and the inner diameter of the chromatographic column is 5 μm;

[0072] Mobile phase gradient: acetonitrile 95% → 60% + 10 mM ammonium acetate pH 4.2, complete carbon chain grouping in 80 min;

[0073] The second dimension is chiral exclusion chromatography:

[0074] The chromatographic column is Chiralcel OD-RH, with a length and inner diameter of 250×2.0 mm and an inner diameter of the column of 3 μm;

[0075] Mobile phase: ethanol-water, with the ratio of ethanol to water being 90:10, containing 0.1% diethylamine, and isocratic elution for separating positional isomers;

[0076] The third dimension is capillary electrophoresis:

[0077] The inner diameter and effective length of the capillary are 50 μmi.d.×50 cm, and the coating is polyethylene glycol;

[0078] The separation voltage is 25 kV and the temperature is 20 °C to achieve the separation of enantiomers with different electrophoretic mobilities;

[0079] The fourth dimension is nano-LC:

[0080] The chromatographic column is Acquity UPLC BEH C18, with a length and inner diameter of 100×0.15 mm and an inner diameter of the column of 1.7 μm;

[0081] Gradient: At 0 minute when injection starts, in the mobile phase, the proportion of methanol is 10%, containing 0.1% formic acid. As time goes by, the proportion of methanol continuously increases and reaches 90% at 15 minutes;

[0082] Step 3: Dynamic mass spectrometry imaging detection: Adopt the mass spectrometry imaging detection technology with dynamic switching of dual ion sources; the dynamic switching period of the dual ion sources ≤ 200 ms, and ≥ 50 mass spectrometry images are collected for each pixel point; the dynamic mass spectrometry imaging detection specifically includes ion source configuration and mass analyzer mode:

[0083] Ion source configuration: atmospheric pressure chemical ionization + electrospray dual source switching;

[0084] Atmospheric pressure chemical ionization ionizes the solvent molecules in the mobile phase through corona discharge, and then collides with the sample molecules to generate ions;

[0085] Electrospray dual source switching atomizes the liquid sample into charged droplets through a high-voltage electric field and generates ions after evaporation;

[0086] Dual source switching 5 times per second: Switching 5 times per second rapidly alternates the acquisition of data from the two ion sources in a single experiment, avoiding manual replacement of hardware;

[0087] The imaging parameters are a spatial resolution of 50 μm: The instrument can distinguish two adjacent points on the sample surface with a distance of 50 μm, ensuring the details of the molecular distribution image; the cumulative time for each pixel point is 100 ms: The signal acquisition time for each pixel point is 0.1 second;

[0088] Mass analyzer mode: Orbitrap resolution 120,000 (m / z 200), mass accuracy error < 1 ppm; linear ion trap for MS 3 fragmentation to obtain the fingerprint of tertiary fragment ions, multi-stage mass spectrometry MS 3 .

[0089] High resolution: A resolution of 120,000 (at m / z 200) means that the instrument can accurately distinguish ions with extremely small differences in mass-to-charge ratio (such as m / z 200.0001 and 200.0002), reducing isotope or background interference.

[0090] Mass accuracy: Error < 1 ppm, which can accurately deduce the molecular formula (for example, the allowable error range for m / z 200 is ±0.0002 Da);

[0091] Step 4. Deep learning data analysis: An overlapping peak analysis algorithm based on a 3D U-Net network; the overlapping peak analysis algorithm based on a 3D U-Net network specifically includes a 3D peak detection algorithm, a background correction model, and an isomer tracing algorithm; among them,

[0092] The 3D peak detection algorithm is based on a U-Net neural network to identify overlapping peaks in the three-dimensional space of chromatography - electrophoresis - mass spectrometry. The training dataset includes the multi-dimensional chromatographic behaviors of more than 1000 LAS isomer standards;

[0093] The input features of the background correction model are retention time, mass-to-charge ratio, isotope peak intensity ratio, and matrix fingerprint; the output is the corrected absolute quantitative concentration, R 2 ≥ 0.995, n = 5000 test set;

[0094] The formula for the isomer tracing algorithm is:

[0095] where I test is the sample ion intensity, and S ref is the standard product spectral library signal;

[0096] Step 5. Precise calculation of isomer content: Input the overlapping peak analysis results obtained from deep learning data analysis, the isomer peak areas separated by each dimension of chromatography, and the standard curve data into a pre-constructed isomer content calculation model. The model uses machine learning algorithms such as multiple linear regression or support vector machine to train and optimize the input data to establish a quantitative relationship between the isomer peak area and content; through model calculation, the precise content of each isomer in LAS is finally output, including relative content and absolute content, providing accurate and reliable data support for subsequent isomer analysis;

[0097] Step 6: Quality Control and Result Verification: Quantitative analysis is carried out using the internal standard method to ensure the accuracy of the data; high-purity standards are used for calibration to reduce errors; blank control experiments are set up to monitor potential contamination during the experiment; at the same time, statistical analysis of the experimental data is performed, including calculating the mean, standard deviation, and coefficient of variation, to evaluate the stability and reliability of the data; in addition, regular participation in external quality assessment programs and comparison with peer laboratories will be carried out to continuously improve the detection level; the specific internal standard method is to add an internal standard with a known concentration to the sample, and perform pretreatment and detection together with the component to be measured, and calculate the content of the component to be measured by comparing the peak area or peak height ratio of the component to be measured and the internal standard.

[0098] Example 1:

[0099] Detection of Marine Microplastics Samples

[0100] Pretreatment Optimization:

[0101] After ultrasonic oscillation elution of microplastic particles (50 - 100μm), MOF materials are used for specific adsorption of LAS isomers;

[0102] Isotope Internal Standard Recovery Rate: 13 The recovery rate of [C6]-2-C12-LAS is 98.7% ± 1.2%, 13 The recovery rate of [C6]-3-C12-LAS is 97.3% ± 1.5%;

[0103] Instrument Parameters:

[0104] The total time of four-stage separation: 120 min, including 40 min in the first dimension + 30 min in the second dimension + 20 min in the third dimension + 30 min in the fourth dimension;

[0105] Mass Spectrometry Imaging Pixel Matrix: 200×200, covering 1 cm 2 Sample surface;

[0106] Detection Results:

[0107]

[0108] Comparative Example: Comparison of Three-stage Separation Methods:

[0109]

[0110]

[0111] In terms of peak capacity, the method adopted in the present invention shows significant advantages, reaching 12,500, which is an 84% increase compared to 6,800 of the traditional three-stage method. This means that under the same separation conditions, the present invention can accommodate more chromatographic peaks, thereby improving the separation efficiency and detection ability of isomers.

[0112] In terms of the detection limit, the method of the present invention also shows a lower detection limit, reaching 12 ppt, while the traditional three-stage method is 85 ppt, and the difference rate is as high as -86%. This means that the present invention has higher sensitivity in the detection of trace isomers and can detect target substances at lower concentrations.

[0113] In addition, in terms of matrix correction error, the method of the present invention is also superior to the traditional three-stage method. The matrix correction error of the present invention is less than or equal to 2.1%, while the error of the traditional method is as high as 7.3% ± 1.5%, and the difference rate is -71%. This further proves the accuracy and reliability of the present invention in complex matrix samples.

[0114] Finally, in terms of the number of isomers resolved, the method of the present invention also shows obvious advantages. By adopting a four-stage chromatographic separation system and deep learning data analysis technology, the present invention can resolve 12 isomers, while the traditional three-stage method can only resolve 7 isomers. This not only increases the types of detected isomers but also provides more information for subsequent isomer analysis and research.

[0115] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A precise detection method for the isomer content in LAS, characterized in that: It includes the following steps: Step 1. Isotope-coded multiplex pretreatment: Specifically, it is divided into seven sub-steps, namely stable isotope labeling, dual-column purification, microwave-assisted derivatization, cryogenic centrifugation for layering, nanomaterial enrichment, on-line solid-phase microextraction, and membrane injection pretreatment; Step 2. Quadrupole chromatography separation system: It includes a quadrupole separation system of hydrophilic interaction-chiral exclusion-capillary electrophoresis-nano-LC; Step 3. Dynamic mass spectrometry imaging detection: It adopts a mass spectrometry imaging detection technology with dynamic switching of dual ion sources; Step 4. Deep learning data analysis: An overlapping peak analysis algorithm based on a 3D U-Net network; Step 5. Precise calculation of isomer content: Input the results obtained from deep learning data analysis into a pre-constructed isomer content calculation model. This model trains and optimizes the analysis data through machine learning algorithms, and finally outputs the precise content of isomers in LAS; Step 6. Quality control and result verification: To ensure the accuracy and reliability of the detection results, a series of quality control measures need to be carried out, including calibration with standards, setting parallel samples for repeatability verification, and cross-verification using third-party methods; at the same time, analyze and correct the errors that may be introduced during the detection process to ensure that the isomer content data in the final report is accurate and error-free.

2. The precise detection method for the isomer content in LAS according to claim 1, characterized in that: The stable isotope labeling in the first step is specifically as follows: Add 10 μL 13 of [C6]-2-C12-LAS internal standard at 100 μg / mL, vortex for 30 s; The dual-column purification is specifically as follows: Pass through a Florisil column and a SAX column in sequence. The Florisil column is used to remove fat-soluble impurities, and the SAX column is used to adsorb anionic interfering substances; The microwave-assisted derivatization is specifically as follows: Add N,O-bis(trimethylsilyl)acetamide (BTSA) and perform microwave treatment at 120 °C for 5 min; The power of the microwave-assisted derivatization step is 300 - 600 W, the time is 4 - 6 min, and the volume ratio of the derivatization reagent to the sample is 1:20 - 1:50; The cryogenic centrifugation for layering is specifically as follows: Freeze at -40 °C for 30 min, centrifuge at 15000 rpm for 20 min, and collect the organic phase; The nanomaterial enrichment is specifically as follows: Add 10 mg of MOF-74-NH2 material, shake and adsorb for 2 h, and elute with methanol three times; The on-line solid-phase microextraction is specifically as follows: The SPME fiber, the solid-phase coating material is PDMS / DVB, the coating thickness is 50 / 30 μm, and headspace extraction is carried out at 70 °C for 15 min; The membrane injection pretreatment is specifically as follows: Filter through a 0.05 μm alumina membrane to remove derivatization by-products.

3. The precise detection method for the isomer content in LAS according to claim 1, characterized in that: The specific description of the quadrupole chromatography separation system in Step 2 is as follows: The first dimension is hydrophilic interaction chromatography: The chromatographic column is Shodex Asahipak HILIC-N, the length and inner diameter of the chromatographic column are 300×4.6 mm, and the inner diameter of the chromatographic column is 5 μm; Mobile phase gradient: Acetonitrile 95% → 60% + 10 mM ammonium acetate pH 4.2, and carbon chain grouping is completed in 80 min; The second dimension is chiral exclusion chromatography: The chromatographic column is Chiralcel OD-RH, the length and inner diameter of the chromatographic column are 250×2.0 mm, and the inner diameter of the chromatographic column is 3 μm; Mobile phase: Ethanol-water, the proportion of ethanol-water is 90:10, containing 0.1% diethylamine, and isocratic elution is used to separate positional isomers; The third dimension is capillary electrophoresis: The inner diameter and effective length of the capillary are 50 μmi.d.×50 cm, and the coating is polyethylene glycol; The separation voltage is 25 kV and the temperature is 20 °C, achieving the separation of isomers based on electrophoretic mobility differences; The fourth dimension is nano-LC: The chromatographic column is Acquity UPLC BEH C18, with a length and inner diameter of 100×0.15 mm and an inner diameter of the column of 1.7 μm; Gradient: At 0 minutes when injection starts, in the mobile phase, the proportion of methanol is 10%, containing 0.1% formic acid. As time goes by, the proportion of methanol continuously increases and reaches 90% at 15 minutes.

4. The precise detection method for the isomer content in LAS according to claim 1, characterized in that: The dynamic mass spectrometry imaging detection in step three specifically includes ion source configuration and mass analyzer mode: Ion source configuration: Atmospheric pressure chemical ionization + electrospray dual-source switching; Atmospheric pressure chemical ionization ionizes the solvent molecules in the mobile phase through corona discharge, and then collides with sample molecules to generate ions; Electrospray dual-source switching atomizes the liquid sample into charged droplets through a high-voltage electric field and generates ions after evaporation; Dual-source switching is 5 times per second: Switching 5 times per second rapidly alternates the acquisition of data from the two ion sources in a single experiment, avoiding manual replacement of hardware; The imaging parameter is a spatial resolution of 50 μm: The instrument can distinguish two adjacent points on the sample surface with a distance of 50 μm, ensuring the details of the molecular distribution image; The cumulative time for each pixel point is 100 ms: The signal acquisition time for each pixel point is 0.1 second; Mass analyzer mode: Orbitrap resolution 120,000 (m / z 200), mass accuracy error < 1 ppm; linear ion trap for MS 3 fragmentation to obtain the fingerprint of the third-stage fragment ions, multi-stage mass spectrometry MS 3 .

5. The precise detection method for the isomer content in LAS according to claim 1, characterized in that: The overlapping peak resolution algorithm based on the three-dimensional U-Net network specifically includes a three-dimensional peak detection algorithm, a background correction model, and an isomer tracing algorithm; among them, The three-dimensional peak detection algorithm is based on the U-Net neural network to identify overlapping peaks in the three-dimensional space of chromatography-electrophoresis-mass spectrometry. The training data set contains the multi-dimensional chromatographic behaviors of more than 1000 LAS isomer standards; The input features of the background correction model are retention time, mass-to-charge ratio, isotope peak intensity ratio, and matrix fingerprint; the output is the corrected absolute quantitative concentration, R2≥0.995, n = 5000 test set; The isomer tracing algorithm formula is as follows: Among them, I test is the ionic strength of the sample, and S ref is the signal of the standard spectral library.

6. The precise detection method for the isomer content in LAS according to claim 1 is characterized in that: In the four-stage chromatographic separation system: The coating material of the third-dimensional capillary electrophoresis is zwitterionic polymer, and the zeta potential ≤ -20 mV; The column oven of the fourth-dimensional nano-LC is equipped with a gradient cooling device, and the temperature fluctuation ≤ ±0.1 °C.

7. The preparation method of a precise detection method for the isomer content in LAS according to claim 1, characterized in that: The dynamic switching period of the dual ion source ≤ 200 ms, and ≥ 50 mass spectrometry images are collected for each pixel point.

8. The preparation method of a precise detection method for the isomer content in LAS according to claim 1, characterized in that: The precise calculation of the isomer content in step five is specifically to input the overlapping peak resolution results obtained from deep learning data analysis, the isomer peak areas separated by each dimension of chromatography, and the standard curve data into a pre-constructed isomer content calculation model. The model uses machine learning algorithms such as multiple linear regression or support vector machine to train and optimize the input data to establish a quantitative relationship between the isomer peak area and the content; through model calculation, the precise content of each isomer in LAS is finally output, including relative content and absolute content, providing accurate and reliable data support for subsequent isomer analysis.

9. The preparation method of a precise detection method for the isomer content in LAS according to claim 1, characterized in that: In step six, the quality control and result verification are specifically as follows: quantitative analysis is carried out using the internal standard method to ensure the accuracy of the data; high-purity standards are used for calibration to reduce errors; blank control experiments are set up to monitor potential contamination during the experiment; at the same time, statistical analysis of the experimental data is performed, including calculating the mean, standard deviation, and coefficient of variation, to evaluate the stability and reliability of the data; in addition, regular participation in external quality assessment programs and comparison with peer laboratories are carried out to continuously improve the detection level; the internal standard method specifically means adding an internal standard with a known concentration to the sample, performing pretreatment and detection on the component to be measured together with the internal standard, and calculating the content of the component to be measured by comparing the peak area or peak height ratio of the component to be measured with that of the internal standard.

Images