Rapid and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method

Abstract

The invention relates to a rapid and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method, which belongs to the technical field of environmental engineering, and comprises the following steps: firstly, continuously passing a water sample to be analyzed through a plurality of solid-phase extraction small columns, and respectively collecting and separating the samples according to column volume multiples; secondly, carrying out ultraviolet visible spectrum analysis on the solid-phase extraction small column separation sample, and simultaneously analyzing the three-dimensional fluorescence spectrum of one sample arranged in the middle according to the amplification sequence in the original water sample and the solid-phase extraction small column outflow amplification sample; then forming a fluorescence quantum yield characteristic matrix of the solid-phase extraction small column separation sample by utilizing the three-dimensional fluorescence spectrum and the ultraviolet visible light spectrum of the amplified sample; reconstructing three-dimensional fluorescence spectrums of all other amplified samples according to the fluorescence quantum yield characteristic matrix and the ultraviolet visible light spectrums; and finally, carrying out parallel factor analysis on the three-dimensional fluorescence spectrums of the amplified sample and the raw water sample to realize rapid and low-cost single-sample three-dimensional fluorescence-parallel factor analysis of the sample.

Description

Technical Field

[0001] The invention belongs to the technical field of environmental engineering, and in particular relates to a fast and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method. Background Art

[0002] As my country continues to pay attention to the water quality safety of key river basins and water bodies, higher requirements are placed on the rapid identification and detection technology of organic pollutants in water pollution incidents. Spectral analysis is widely used in the identification and detection of organic matter in water. Among them, fluorescence spectroscopy has the characteristics of simple pre-treatment, rapidity, and high sensitivity in identifying organic pollutants in water. Three-dimensional fluorescence spectroscopy can contain the fingerprint information of dissolved organic matter in water, and its application is even more extensive. However, the fluorophores of various organic matter in the three-dimensional fluorescence spectrum overlap and stack with each other, making it impossible for the three-dimensional fluorescence spectrum to identify organic matter in water. The parallel factor method is a technology that uses mathematical methods to separate overlapping fluorophores based on the trilinear principle. It is widely used in three-dimensional fluorescence spectrum analysis and organic matter analysis.

[0003] However, traditional three-dimensional fluorescence-parallel factor analysis requires the sample size to be as large as possible and to be independent of each other in order to ensure the stability of the model. However, in practical applications, the collection of large-scale samples requires a lot of time, making three-dimensional fluorescence-parallel factor analysis impossible to use in the rapid identification of organic matter in water. Single-sample three-dimensional fluorescence-parallel factor analysis can solve the above problems. However, according to previous studies and patent searches, single-sample three-dimensional fluorescence-parallel factor analysis requires the use of large-scale analytical instruments or the use of resins to expand the three-dimensional fluorescence spectrum sample size. Large-scale analytical instruments or three-dimensional fluorescence spectrum scanning of expanded samples will increase the analysis cost and time consumption of the samples to be tested.

[0004] To this end, the present invention proposes a fast and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method, which can realize the rapid analysis, identification and traceability of dissolved organic matter in water. Summary of the Invention

[0005] In view of the shortcomings of the existing technology, the present invention proposes a fast and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method to solve or improve the above problems.

[0006] The present invention adopts the following technical solutions to solve the above problems:

[0007] A rapid and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method, comprising the following steps:

[0008] S1: Take a water sample from the target water body to be analyzed, remove suspended matter using a filter membrane to obtain a water sample, and take a portion of the water sample to analyze its three-dimensional fluorescence spectrum;

[0009] S2: amplifying the water sample obtained in S1 using a solid phase extraction cartridge to obtain an amplified sample;

[0010] S3: analyzing the UV-visible spectrum of the amplified sample;

[0011] S4: analyzing the three-dimensional fluorescence spectrum of a sample arranged in the middle of the elution order among the amplified samples flowing out of each solid phase extraction cartridge;

[0012] S5: forming a fluorescence quantum yield characteristic matrix of each solid phase extraction cartridge amplified sample of the water sample according to the ultraviolet-visible light spectrum and three-dimensional fluorescence spectrum obtained in S3 and S4;

[0013] S6: reconstructing the three-dimensional fluorescence spectra of all other amplified samples using the fluorescence quantum yield characteristic matrix and the ultraviolet-visible light spectrum of the amplified sample;

[0014] S7: performing parallel factor analysis on the three-dimensional fluorescence data of the water sample and the amplified sample to analyze the stability of the model and obtain component load information of the abnormal sample and score information of each component of the sample.

[0015] Furthermore, in S1 and S4, the three-dimensional fluorescence spectra of the water sample and the amplified sample need to be preprocessed by using the Delaunay triangle interpolation method to eliminate Rayleigh and Raman scattering to reduce their interference with the sample organic matter fluorophores.

[0016] Furthermore, after pre-processing the three-dimensional fluorescence spectra of the water sample and the amplified sample, the three-dimensional fluorescence spectra are standardized using Raman peak standardization at an excitation wavelength of 350 nm to achieve universal data between different instruments.

[0017] Furthermore, in S2, the preparation of the amplified sample specifically comprises the following steps:

[0018] S201: amplifying the water sample to be analyzed using a solid phase extraction cartridge, and selecting a corresponding number and type of solid phase extraction cartridges for amplification according to the source of the water sample;

[0019] S202: The sample amplification process is: the water sample is passed through a plurality of solid phase extraction cartridges continuously and separately, and the amplified samples passing through the solid phase extraction cartridges are continuously collected, and the amount of the amplified samples collected each time is determined according to the volume of the solid phase extraction cartridge.

[0020] Furthermore, in 201, if the water sample originates from the secondary effluent of a sewage treatment plant or the receiving water body of a sewage treatment plant, four solid phase extraction cartridges, namely C8, NH2, PPL and HLB, are used to amplify the sample; if the water sample originates from a natural water body, three solid phase extraction cartridges, namely PPL, NH2 and SAX, are used to amplify the sample.

[0021] Furthermore, in S3, when analyzing the ultraviolet-visible spectrum of the amplified water sample, the ultraviolet-visible spectrum scanning wavelength range is consistent with the three-dimensional fluorescence spectrum excitation wavelength range of the sample.

[0022] Furthermore, S5 specifically includes the following steps:

[0023] S501: According to the calculation formula of the three-dimensional fluorescence spectrum intensity:

[0024]

[0025] Among them, η(λ ex ,λ em ) is the three-dimensional fluorescence spectrum of organic matter at the excitation wavelength λ ex , the emission wavelength is λ em The fluorescence quantum yield coefficient when F(λ ex ,λ em ) is the three-dimensional fluorescence spectrum of organic matter at the excitation wavelength λ ex , the emission wavelength is λ em The fluorescence intensity at the time of F0(λ ex ) and Abs(λ ex ) are the three-dimensional fluorescence spectra of organic matter at the excitation wavelength λ ex When the incident light intensity and absorbance are equal, K is a constant;

[0026] When Abs(λ ex )<0.02, formula (1) can be approximated as:

[0027] F(λ ex ,λ em )=Kη(λ ex ,λ em )F0(λ ex )[Abs(λ ex )] (2)

[0028] S502: The characteristic matrix of the fluorescence quantum yield of organic matter can be expressed as:

[0029]

[0030] S503: After the three-dimensional fluorescence spectrum is normalized by Raman peak, F0(λ ex ) is a constant; Abs(λex ) is the incident light intensity of the fluorescence spectrometer F0(λ ex ) is not measurable, so the absorbance value of the UV-visible spectrophotometer is used as a substitute and multiplied by a coefficient:

[0031] Abs(λ ex )=T·Abs′(λ ex ) (4)

[0032] Among them, Abs(λ ex ) is the incident light intensity of the fluorescence spectrometer F0(λ ex ) absorbance; Abs'(λ ex ) is the absorbance of the alternative UV-visible spectrophotometer; T is the conversion coefficient (constant);

[0033] S504: The characteristic matrix of the fluorescence quantum yield of organic matter can be further expressed as:

[0034]

[0035] S505: In order to prevent Abs'(λ ex ) is close to 0, which causes η' to increase abnormally (singular point), and a correction factor M is introduced:

[0036]

[0037] Among them, M(λ ex ,λ em ) is the organic matter at the excitation wavelength λ ex , the emission wavelength is λ em Correction coefficient when F max and F min is the maximum and minimum fluorescence intensity in the three-dimensional fluorescence matrix;

[0038] S506: Formula (5) can be expressed as:

[0039]

[0040] Therefore, the fluorescence quantum yield characteristic matrix of organic matter in water samples can be obtained from formula (7).

[0041] Furthermore, S6 specifically includes the following steps:

[0042] Substitute the UV-visible light spectrum absorbance value of the amplified sample into formula (2) and formula (7) and arrange them to obtain:

[0043]

[0044] Among them, F'(λ ex ,λem ) is the reconstructed three-dimensional fluorescence spectrum of organic matter at the excitation wavelength λ ex , the emission wavelength is λ em Abs" (λ ex ) is the UV-visible absorbance of the amplified sample.

[0045] Furthermore, S7 specifically includes the following steps:

[0046] S701: merging the three-dimensional fluorescence data of the original water sample and the amplified sample to obtain i groups of three-dimensional fluorescence data, and composing the three-dimensional fluorescence data of the i samples into an M*N*I three-dimensional data matrix;

[0047] S702: Using the trilinear principle of the parallel factor method, the matrix X is decomposed into F factors. Then the original matrix X can be expressed as:

[0048]

[0049] Among them, F represents the number of factors, a m,f represents the mth element in the vector af, a m,f 、b n,f and c i,f Construct matrices A, B, C, x respectively m,n,i (m=1,…,M,n=1,…,N,i=1,…,I) is the M*N*I three-dimensional data matrix X, e m,n,i (m=1,…,M,n=1,…,N,i=1,…,I) is the three-dimensional error matrix E of M*N*I;

[0050] S703: Decompose the matrix X into three matrices A, B, and C, which are respectively the score matrix (i.e., the relative mass concentration matrix C of each component), the load matrix A corresponding to the emission spectrum, and the load matrix B corresponding to the excitation spectrum;

[0051] S704: According to formula (9), set multiple F values, and iterate and calculate A for each F value T 、B T 、C T The least squares estimate is performed until E converges to the minimum and the kernel consistency is the highest. At this time, the F value is the optimal value. At this time, the three-dimensional fluorescence data of the sample set can be decomposed into F factors according to the parallel factor method, and the parallel factor model at this time is the optimal model.

[0052] Advantages of the present invention:

[0053] The present invention provides a rapid and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method. By utilizing a common low-cost solid-phase extraction cartridge to amplify the sample, a rapid and low-cost construction of the single-sample three-dimensional fluorescence-parallel factor analysis method is achieved. By analyzing the fluorescence quantum yield characteristic matrix of each solid-phase extraction cartridge-amplified sample, the three-dimensional fluorescence spectrum is reconstructed using the ultraviolet-visible spectrum of the amplified sample, thereby avoiding performing three-dimensional fluorescence spectrum analysis on each amplified sample and improving the analysis speed of the method. By improving the correction factor M, the weakening of details in the low quantum yield region in the light quantum characteristic matrix caused by eliminating singular points using Wiener filtering is avoided. The present method can realize rapid analysis, identification and traceability of soluble organic matter in water. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The exemplary embodiments of the present invention will be described in more detail with reference to the accompanying drawings, so that the above-mentioned objects, features and advantages of the present invention will become more apparent. In the exemplary embodiments of the present invention, the same reference numerals generally represent the same components.

[0055] Figure 1 It is a technical flow chart of the present invention;

[0056] Figure 2 The three-dimensional fluorescence spectrum of the water sample of Qiangzi River in Example 2;

[0057] Figure 3 (a) is the UV-visible spectrum of a sample in the middle of the elution order of the water sample from Qiangzi River in Example 2 after passing through the NH2 solid phase extraction cartridge;

[0058] Figure 3 (b) is the three-dimensional fluorescence spectrum of a sample in the middle of the elution order of the water sample from Qiangzi River in Example 2 passing through the NH2 solid phase extraction cartridge;

[0059] Figure 4 The three-dimensional fluorescence spectrum reconstructed from the amplified sample of the NH2 solid phase extraction cartridge in Example 2;

[0060] Figure 5 This is the component diagram of the single sample three-dimensional fluorescence-parallel factor analysis in Example 2;

[0061] Figure 6 This is a component diagram of traditional three-dimensional fluorescence-parallel factor analysis. DETAILED DESCRIPTION

[0062] The specific embodiments of the present invention are described below to facilitate understanding of the present invention by those skilled in the art. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, the main various changes are within the spirit and scope of the present invention as defined and determined by the appended claims. These changes are obvious, and all inventions and creations utilizing the concepts of the present invention are protected.

[0063] It should be noted that the professional terms used in the present invention are only for the purpose of describing specific embodiments and are not intended to limit the scope of protection of the present invention. Unless otherwise specified, the various raw materials, reagents, instruments and equipment used in the following embodiments of the present invention can be purchased from the market or prepared by existing methods.

[0064] Example 1

[0065] According to an embodiment of the present application, the specific process is as follows: Figure 1 This protocol provides a rapid and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method, which includes the following steps:

[0066] S1: Take a water sample from the target water body to be analyzed, use a filter membrane to remove suspended matter to obtain a water sample, and take a portion of the water sample to analyze its three-dimensional fluorescence spectrum.

[0067] The three-dimensional fluorescence spectra of water samples need to be preprocessed. Rayleigh and Raman scattering can be eliminated through Delaunay triangle interpolation method to reduce their interference with the fluorophores of organic matter in the sample. After preprocessing the three-dimensional fluorescence spectra of water samples and amplified samples, the three-dimensional fluorescence spectra are standardized using Raman peak standardization at an excitation wavelength of 350nm to achieve universal data between different instruments.

[0068] S2: Amplify the water sample obtained in S1 using a solid phase extraction cartridge to obtain an amplified sample.

[0069] S201: Amplifying the water sample to be analyzed using a solid phase extraction cartridge, wherein a corresponding number and type of solid phase extraction cartridges are selected for amplification according to the source of the water sample;

[0070] S202: The sample amplification process is: the water sample is passed through a plurality of solid phase extraction cartridges continuously and separately, and the amplified samples passing through the solid phase extraction cartridges are continuously collected, and the amount of the amplified samples collected each time is determined according to the volume of the solid phase extraction cartridge.

[0071] S3: Analyze the UV-visible spectrum of the amplified sample.

[0072] When analyzing the ultraviolet-visible spectrum of the amplified water sample, the ultraviolet-visible spectrum scanning wavelength range is consistent with the three-dimensional fluorescence spectrum excitation wavelength range of the sample.

[0073] S4: Analyze the three-dimensional fluorescence spectrum of the sample in the middle of the amplified samples flowing out of each solid phase extraction cartridge in the order of elution.

[0074] The three-dimensional fluorescence spectrum of the amplified sample needs to be preprocessed. Rayleigh and Raman scattering are eliminated through Delaunay triangle interpolation method to reduce their interference with the fluorophores of organic matter in the sample; the three-dimensional fluorescence spectrum is standardized using the Raman peak standardization at an excitation wavelength of 350nm to achieve data commonality between different instruments.

[0075] S5: Based on the UV-visible light spectrum and three-dimensional fluorescence spectrum obtained in S3 and S4, a fluorescence quantum yield characteristic matrix of each solid phase extraction cartridge amplified sample of the water sample is formed, which specifically includes the following steps:

[0076] S501: According to the calculation formula of the three-dimensional fluorescence spectrum intensity:

[0077]

[0078] Among them, η(λ ex ,λ em ) is the three-dimensional fluorescence spectrum of organic matter at the excitation wavelength λ ex , the emission wavelength is λ em The fluorescence quantum yield coefficient when F(λ ex ,λ em ) is the three-dimensional fluorescence spectrum of organic matter at the excitation wavelength λ ex , the emission wavelength is λ em The fluorescence intensity at the time of F0(λ ex ) and Abs(λ ex ) are the three-dimensional fluorescence spectra of organic matter at the excitation wavelength λ ex When the incident light intensity and absorbance are equal, K is a constant;

[0079] When Abs(λ ex )<0.02, formula (1) can be approximated as:

[0080] F(λ ex ,λ em )=Kη(λ ex ,λ em )F0(λ ex )[Abs(λ ex )] (2)

[0081] S502: The characteristic matrix of the fluorescence quantum yield of organic matter can be expressed as:

[0082]

[0083] S503: After the three-dimensional fluorescence spectrum is normalized by Raman peak, F0(λ ex ) is a constant; Abs(λ ex ) is the incident light intensity of the fluorescence spectrometer F0(λ ex ) is not measurable, so the absorbance value of the UV-visible spectrophotometer is used as a substitute and multiplied by a coefficient:

[0084] Abs(λ ex )=T·Abs′(λ ex ) (4)

[0085] Among them, Abs(λ ex ) is the incident light intensity of the fluorescence spectrometer F0(λ ex ) absorbance; Abs'(λ ex ) is the absorbance of the alternative UV-visible spectrophotometer; T is the conversion coefficient (constant);

[0086] S504: The characteristic matrix of the fluorescence quantum yield of organic matter can be further expressed as:

[0087]

[0088] S505: In order to prevent Abs'(λ ex ) is close to 0, which causes η' to increase abnormally (singular point), and a correction factor M is introduced:

[0089]

[0090] Among them, M(λ ex ,λ em ) is the organic matter at the excitation wavelength λ ex , the emission wavelength is λ em Correction coefficient when F max and F min is the maximum and minimum fluorescence intensity in the three-dimensional fluorescence matrix;

[0091] S506: Formula (5) can be expressed as:

[0092]

[0093] Therefore, the fluorescence quantum yield characteristic matrix of organic matter in water samples can be obtained from formula (7).

[0094] S6: Reconstruct the three-dimensional fluorescence spectra of all other amplified samples using the fluorescence quantum yield characteristic matrix and the UV-visible light spectrum of the amplified sample.

[0095] Substitute the UV-visible light spectrum absorbance value of the amplified sample into formula (2) and formula (7) and arrange them to obtain:

[0096]

[0097] Among them, F'(λ ex ,λ em ) is the reconstructed three-dimensional fluorescence spectrum of organic matter at the excitation wavelength λ ex , the emission wavelength is λ em Abs" (λ ex ) is the UV-visible absorbance of the amplified sample.

[0098] S7: Performing parallel factor analysis on the three-dimensional fluorescence data of the water sample and the amplified sample to analyze the stability of the model and obtain the component loading information and score information of each component of the abnormal sample, which specifically includes the following steps:

[0099] S701: merging the three-dimensional fluorescence data of the original water sample and the amplified sample to obtain i groups of three-dimensional fluorescence data, and composing the three-dimensional fluorescence data of the i samples into an M*N*I three-dimensional data matrix;

[0100] S702: Using the trilinear principle of the parallel factor method, the matrix X is decomposed into F factors. Then the original matrix X can be expressed as:

[0101]

[0102] Among them, F represents the number of factors, a m,f represents the mth element in the vector af, a m,f 、b n,f and c i,f Construct matrices A, B, C, x respectively m,n,i (m=1,…,M,n=1,…,N,i=1,…,I) is the M*N*I three-dimensional data matrix X, e m,n,i (m=1,…,M,n=1,…,N,i=1,…,I) is the three-dimensional error matrix E of M*N*I;

[0103] S703: Decompose the matrix X into three matrices A, B, and C, which are respectively the score matrix (i.e., the relative mass concentration matrix C of each component), the load matrix A corresponding to the emission spectrum, and the load matrix B corresponding to the excitation spectrum;

[0104] S704: According to formula (9), set multiple F values, and iterate and calculate A for each F value T 、B T 、C TThe least squares estimate is performed until E converges to the minimum and the kernel consistency is the highest. At this time, the F value is the optimal value. At this time, the three-dimensional fluorescence data of the sample set can be decomposed into F factors according to the parallel factor method, and the parallel factor model at this time is the optimal model.

[0105] Example 2

[0106] This example takes water samples from Qiangzi River, which is the receiving water body of the sewage treatment plant, and specifically includes the following steps:

[0107] (1) Analyze the three-dimensional fluorescence spectrum of the sample, with the excitation wavelength range from 200 to 450 nm, the emission wavelength range from 250 to 550 nm, the step length of 5 nm, the slit width of 5 nm, and the scanning speed of 1200 nm / min. The three-dimensional fluorescence spectrum of the water sample of Qiangzi River is obtained, as shown in Figure 2 As shown in the figure, the three-dimensional fluorescence spectrum of the Qiangzi River water sample mainly shows that the sample contains tryptophan (Ex / Em=200-250nm / 330-400nm), soluble microbial products (Ex / Em=250-300nm / 310-370nm), humic acid (Ex / Em=250-370nm / 370-500nm) and fulvic acid organic matter (Ex / Em=200-250nm / 370-500nm);

[0108] (2) Qiangzi River is the receiving water body of the sewage treatment plant, so four solid phase extraction cartridges, C8, NH2, PPL, and HLB, were used to amplify the samples. The solid phase extraction cartridges filtered 5 mL of methanol and 5 mL of ultrapure water in sequence for activation. Each cartridge filtered 45 mL of the original sample, and each amplified sample collected 3 mL. A total of 15 amplified samples were collected in each cartridge.

[0109] (3) measuring the UV-visible spectra of all amplified samples, scanning the wavelength range from 200 to 450 nm with a step size of 5 nm;

[0110] (4) selecting the UV-visible spectrum of an amplified sample in the middle of the elution order of the solid phase extraction cartridge and analyzing the three-dimensional fluorescence spectrum of the sample;

[0111] like Figure 3 As shown, taking NH2 solid phase extraction cartridge as an example, Figure 3 (a) shows the UV-visible spectrum of a sample from the Qiangzi River passing through an NH2 solid-phase extraction cartridge in the middle of the elution order. The high absorbance (shoulder absorption) in the 200-220 nm wavelength range is mainly related to the inorganic salt content in the water. The peak absorption in the 220-250 nm wavelength range indicates that the organic matter in the sample mainly contains unsaturated double bonds or aromatic structures. Figure 3(b) shows the three-dimensional fluorescence spectrum of a sample in the middle of the elution order of the Qiangzi River water sample passing through the NH2 solid phase extraction column, indicating that the sample mainly contains tryptophan (Ex / Em=200-250nm / 330-400nm), soluble microbial products (Ex / Em=250-300nm / 310-370nm), humic acid (Ex / Em=250-370nm / 370-500nm) and fulvic acid (Ex / Em=200-250nm / 370-500nm) organic matter. The fluorescence intensity of these organic matter is lower than that of the organic matter in the raw water ( Figure 2 ), indicating that some organic matter was adsorbed by the NH2 solid phase extraction column, and different components of the organic matter were separated to different degrees;

[0112] (5) forming a fluorescence quantum yield characteristic matrix of the amplified samples effluent from the solid phase extraction cartridge based on the three-dimensional fluorescence spectrum and ultraviolet-visible light spectrum of a sample arranged in the middle of the amplified samples effluent from the solid phase extraction cartridge;

[0113] (6) Using the fluorescence quantum yield characteristic matrix and the UV-visible spectrum of the amplified sample, reconstruct the three-dimensional fluorescence spectra of all other amplified samples, where, Figure 4 As shown in the figure, taking the amplified sample eluted from the NH2 solid phase extraction cartridge as an example, the reconstructed three-dimensional fluorescence spectrum of the amplified sample is shown. The reconstructed three-dimensional fluorescence spectrum shows that the amplified samples contain organic components such as tryptophan (Ex / Em=200-250nm / 330-400nm), soluble microbial products (Ex / Em=250-300nm / 310-370nm), humic acid (Ex / Em=250-370nm / 370-500nm) and fulvic acid (Ex / Em=200-250nm / 370-500nm), but the fluorescence intensities are different, which increases the variability of the fluorescence components. This is the main reason why the subsequent parallel factor analysis can successfully model;

[0114] (7) Perform parallel factor analysis on the three-dimensional fluorescence data of the original sample and the amplified sample, and obtain the component information of the sample. Figure 5 It is the component diagram of single sample three-dimensional fluorescence-parallel factor analysis of water sample. Figure 6 This is a traditional three-dimensional fluorescence-parallel factor analysis component diagram. The results are shown in Table 1:

[0115] Table 1

[0116]

[0117] Table 1 shows the corresponding component information and cosine similarity of single-sample three-dimensional fluorescence-parallel factor analysis and traditional three-dimensional fluorescence-parallel factor analysis. The cosine similarity of the corresponding components is greater than 0.95, indicating the effectiveness of the single-sample three-dimensional fluorescence-parallel factor analysis proposed in the present invention.

[0118] The present invention has been described in detail above through the embodiments. However, the contents are only preferred embodiments of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A rapid and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method, characterized by: The method comprises the following steps: S1: Take a water sample from the target water body to be analyzed, remove suspended matter using a filter membrane to obtain a water sample, and take a portion of the water sample to analyze its three-dimensional fluorescence spectrum; S2: amplifying the water sample obtained in S1 using a solid phase extraction cartridge to obtain an amplified sample; S3: analyzing the UV-visible spectrum of the amplified sample; S4: analyzing the three-dimensional fluorescence spectrum of a sample arranged in the middle of the elution order among the amplified samples flowing out of each solid phase extraction cartridge; S5: forming a fluorescence quantum yield characteristic matrix of each solid phase extraction cartridge amplified sample of the water sample according to the ultraviolet-visible light spectrum and three-dimensional fluorescence spectrum obtained in S3 and S4; S6: reconstructing the three-dimensional fluorescence spectra of all other amplified samples using the fluorescence quantum yield characteristic matrix and the ultraviolet-visible light spectrum of the amplified sample; S7: performing parallel factor analysis on the three-dimensional fluorescence data of the water sample and the amplified sample to analyze the stability of the model and obtain component load information of the abnormal sample and score information of each component of the sample.

2. A rapid and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method according to claim 1, characterized in that: In S1 and S4, the three-dimensional fluorescence spectra of the water sample and the amplified sample need to be preprocessed by using the Delaunay triangle interpolation method to eliminate Rayleigh and Raman scattering to reduce their interference with the sample organic matter fluorophores.

3. A rapid and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method according to claim 2, characterized in that: After pre-processing the three-dimensional fluorescence spectra of the water sample and the amplified sample, the three-dimensional fluorescence spectra are standardized using Raman peak standardization at an excitation wavelength of 350 nm to achieve universality of data between different instruments.

4. A rapid and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method according to claim 1, characterized in that: In S2, the preparation of the amplification sample specifically comprises the following steps: S201: amplifying the water sample to be analyzed using a solid phase extraction cartridge, and selecting a corresponding number and type of solid phase extraction cartridges for amplification according to the source of the water sample; S202: The sample amplification process is: the water sample is passed through a plurality of solid phase extraction cartridges continuously and separately, and the amplified samples passing through the solid phase extraction cartridges are continuously collected, and the amount of the amplified samples collected each time is determined according to the volume of the solid phase extraction cartridge.

5. A rapid and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method according to claim 4, characterized in that: In S201, if the water sample originates from the secondary effluent of a sewage treatment plant or the receiving water body of a sewage treatment plant, four solid phase extraction cartridges, namely C8, NH2, PPL and HLB, are used to amplify the sample; if the water sample originates from a natural water body, three solid phase extraction cartridges, namely PPL, NH2 and SAX, are used to amplify the sample.

6. A rapid and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method according to claim 1, characterized in that: In S3, when analyzing the ultraviolet-visible light spectrum of the amplified water sample, the ultraviolet-visible light spectrum scanning wavelength range is consistent with the three-dimensional fluorescence spectrum excitation wavelength range of the sample.

7. A rapid and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method according to claim 1, characterized in that: S5 specifically includes the following steps: S501: According to the calculation formula of the three-dimensional fluorescence spectrum intensity: Among them, η(λ ex ,λ em ) is the three-dimensional fluorescence spectrum of organic matter at the excitation wavelength λ ex , the emission wavelength is λ em The fluorescence quantum yield coefficient when F(λ ex ,λ em ) is the three-dimensional fluorescence spectrum of organic matter at the excitation wavelength λ ex , the emission wavelength is λ em The fluorescence intensity at the time of F0(λ ex ) and Abs(λ ex ) are the three-dimensional fluorescence spectra of organic matter at the excitation wavelength λ ex When the incident light intensity and absorbance are equal, K is a constant; When Abs(λ ex )<0.02, formula (1) can be approximated as: S502: The characteristic matrix of the fluorescence quantum yield of organic matter can be expressed as: S503: After the three-dimensional fluorescence spectrum is normalized by Raman peak, F0(λ ex ) is a constant; Abs(λ ex ) is the incident light intensity of the fluorescence spectrometer F0(λ ex ) is not measurable, so the absorbance value of the UV-visible spectrophotometer is used as a substitute and multiplied by a coefficient: Abs(λ ex )=T·Abs′(λ ex ) (4) Among them, Abs(λ ex ) is the incident light intensity of the fluorescence spectrometer F0(λ ex ) absorbance; Abs'(λ ex ) is the absorbance of the alternative UV-visible spectrophotometer; T is the conversion coefficient (constant); S504: The characteristic matrix of the fluorescence quantum yield of organic matter can be further expressed as: S505: In order to prevent Abs'(λ ex ) is close to 0, which causes η' to increase abnormally (singular point), and a correction factor M is introduced: Among them, M(λ ex ,λ em ) is the organic matter at the excitation wavelength λ ex , the emission wavelength is λ em Correction coefficient when F max and F min is the maximum and minimum fluorescence intensity in the three-dimensional fluorescence matrix; S506: Formula (5) can be expressed as: Therefore, the fluorescence quantum yield characteristic matrix of organic matter in water samples can be obtained from formula (7).

8. The rapid and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method according to claim 1, characterized in that: S6 specifically includes the following steps: Substitute the UV-visible light spectrum absorbance value of the amplified sample into formula (2) and formula (7) and arrange them to obtain: Among them, F'(λ ex ,λ em ) is the reconstructed three-dimensional fluorescence spectrum of organic matter at the excitation wavelength λ ex , the emission wavelength is λ em Abs" (λ ex ) is the UV-visible absorbance of the amplified sample.

9. The rapid and low-cost single-sample three-dimensional fluorescence-parallel factor analysis method according to claim 1, characterized in that: S7 specifically includes the following steps: S701: merging the three-dimensional fluorescence data of the original water sample and the amplified sample to obtain i groups of three-dimensional fluorescence data, and composing the three-dimensional fluorescence data of the i samples into an M*N*I three-dimensional data matrix; S702: Using the trilinear principle of the parallel factor method, the matrix X is decomposed into F factors. Then the original matrix X can be expressed as: Among them, F represents the number of factors, a m,f represents the mth element in the vector af, a m,f 、b n,f and c i,f Construct matrices A, B, C, x respectively m,n,i (m=1,…,M,n=1,…,N,i=1,…,I) is the M*N*I three-dimensional data matrix X, e m,n,i (m=1,…,M,n=1,…,N,i=1,…,I) is the three-dimensional error matrix E of M*N*I; S703: Decompose the matrix X into three matrices A, B, and C, which are respectively the score matrix (i.e., the relative mass concentration matrix C of each component), the load matrix A corresponding to the emission spectrum, and the load matrix B corresponding to the excitation spectrum; S704: According to formula (9), set multiple F values, and iterate and calculate A for each F value T 、B T 、C T The least squares estimate is performed until E converges to the minimum and the kernel consistency is the highest. At this time, the F value is the optimal value. At this time, the three-dimensional fluorescence data of the sample set can be decomposed into F factors according to the parallel factor method, and the parallel factor model at this time is the optimal model.

Images