CN105699345A - Method for measuring pollutants by virtue of combination of three-dimensional fluorescence spectrum and PARAFAC algorithm - Google Patents

Abstract

The invention provides a method for measuring pollutants by virtue of combination of a three-dimensional fluorescence spectrum and a PARAFAC algorithm, in particular to a method for rapidly identifying pollutants with fluorescence characteristics, such as hydrophobic organic contaminants with conjugated structures, benzene, phenanthrene, naphthalene, anthracene and derivatives thereof, benzoapyrene, 1-naphthol and 2-naphthol, in environments such as water and soil. According to the method, a sample data library is established in advance, and the correlation coefficient is more than 0.998; the method has the advantages of high accuracy, high sensitivity, good repeatability, simplicity and convenience in operation, low analysis cost and the like, and is capable of realizing rapid identification on organic pollutants in environments such as water and soil.

Description

A kind of method that three-dimensional fluorescence spectrum measures pollutant in conjunction with PARAFAC algorithm

Technical field

The invention provides a kind of method that three-dimensional fluorescence spectrum measures pollutant in conjunction with PARAFAC algorithm, relate to pollutant kind and content differentiates field, particularly quickly differentiate the organic pollution in the environment such as water body, soil, as having the hydrophobic organic pollutant of conjugated structure, benzene, phenanthrene, naphthalene, anthracene and derivant, benzo pyrrole, 1-naphthols, the method for the pollutant such as beta naphthal。

Background technology

Fluorescence analysis method is one of important means of routine analysis, but, along with the complexity day by day analyzing the development of object, analysis task, traditional fluorimetry has been difficult to meet requirement。Because traditional fluorescent emission (excites) spectrum simply to excite at some and scans under (transmitting) wavelength, and fluorescence is excitation wavelength and the function launching wavelength, so traditional fluorescent emission (or exciting) spectrum can not be fully described by the fluorescent characteristics of material, therefore one compound fluorescence information of complete description needs three-dimensional fluorescence spectrum to realize, and this is also by the essential condition of spectral matching factor, sign。Additionally, the object that one fluorescence spectrum (launch/excite) containing various ingredients is overlapping, traditional peak value quantitative method is difficult to solve the interference problem between component, need from object more completely information, find selective region, or be only possible in conjunction with other optimization means realize multiple component exactly while analyze。It is the new fluorescence analysis grown up for nearly more than 30 years that three-dimensional fluorescence spectrum analyzes method, this technology is obtained in that fluorescence intensity information when excitation wavelength changes simultaneously with transmitting wavelength or other variable, and fluorescence intensity is expressed as excitation wavelength-transmitting wavelength or the function of two variablees such as wavelength-time, wavelength-phase angle。

Generally, three dimensions of three-dimensional fluorescence refer to fluorescence excitation, launch wavelength and fluorescence intensity, and its performance is that fluorescence intensity is with the information exciting and launching wavelength to change simultaneously。The general method obtaining three-dimensional fluorescence data is to scan emission spectrum on different excitation wavelength positions continuously, and it is showed by available various mapping softwares with Enhanced Capability of Graphic Display And Automatic Judgment forms such as isogonism three-dimensional fluorescence projection (Ex-Em-If) or equal pitch contour spectrum (Ex-Em)。

Further, when processing the system of multicomponent system or component Overlapping of fluorescence spectra, traditional fluorescence analysis utilizes the peak fluorescence intensity of fluorescent material to carry out quantitatively, and its selectivity, sensitivity are subjected to very big restriction。Therefore, it is quite necessary to find a kind of method quick, convenient, stable, mensuration pollutant reliably。

Summary of the invention

In order to solve above-mentioned technical problem, the invention provides a kind of method that three-dimensional fluorescence spectrum measures pollutant in conjunction with PARAFAC algorithm, particularly quickly differentiate the pollutant having fluorescent characteristic in the environment such as water body, soil, as having the hydrophobic organic pollutant of conjugated structure, benzene, phenanthrene, naphthalene, anthracene and derivant, benzo pyrrole, 1-naphthols, the method for the pollutant such as beta naphthal。

In order to realize the technical purpose of the present invention, the present invention adopts the following technical scheme that。

A kind of method that three-dimensional fluorescence spectrum measures pollutant in conjunction with PARAFAC algorithm, described method differentiates that step is as follows:

A. the foundation of sample database: initially set up variable concentrations, the different types of pollutant three-dimensional fluorescence spectrum PARAFAC algorithm data-base in particular excitation/emission wavelength ranges, obtain the standard curve of pollutant levels and Characteristic fluorescence intensity simultaneously；

B. sample pre-treatments: be dissolved in deionized water by sample, passes into a certain amount of nitrogen and removes dissolved oxygen then the filtering with microporous membrane with 0.45 μm, remove impurity；

C. the acquisition of three-dimensional fluorescence spectrum data: the sample after removing impurity in step b is taken in 4ml addition cuvette, is placed in three-dimensional fluorescence spectrum analyser, sets parameter, be scanned obtaining the three-dimensional fluorescence spectrum data of sample；

D. the discriminating of pollutant type: the sample three-dimensional fluorescence spectrum data matlab software obtained in step c is loaded in the workbox equipped with PARAFAC algorithm, spectrum is resolved, position according to fluorescence spectral characteristic peak, in conjunction with the three-dimensional fluorescence spectrum data base of pollutant, differentiates the type of pollutant；

E. pollutant levels differentiate: by the pollutant after the determination type that obtains in step d, contrast in the fluorescence intensity of particular excitation/transmitted wave strong point and the intensity of step a standard curve, it is thus achieved that concentration data；

F. Modifying model: the three-dimensional fluorescence spectrum intensity of target contaminant is beyond the detection upper lower limit value of three-dimensional fluorescence spectrum data base, now, the concentration correspondence of target contaminant is diluted or concentration x times, then through step c, d, e, thus corresponding calculating obtains meeting the target pollutant concentration of three-dimensional fluorescence spectrum standard database comparison。

Further, in described step d, PARAFAC algorithm is: the whole fluorescence datas obtained are built into three-dimensional data matrix X_ijk, three-dimensional data matrix X_ijkIt is decomposed into 3 loading matrixs and residual error number gust, such as drag:

$X_{ijk}={\mathrm{\Σ}}_{f=1}^F{a}_{!f}{b}_{jf}{c}_{kf}+{\mathrm{\ϵ}}_{ijk};i=1,\mathrm{....},I;j=1,\mathrm{...},J;k=1,\mathrm{...},K$

X in formula_ijkAt excitation wavelength j and the wavelength k fluorescence intensity data matrix constituted is launched for all samples i；F is the independent fluorescence number of components that system has actual contribution；A_ifFor the relative concentration battle array of f fluorophor in i-th sample；B_jfAnd c_kfTransmitting that respectively f fluorophor is corresponding and excitation spectrum battle array；ε_ijkFor the residual error battle array of model, solving of it is based on the alternately principle of least square and minimize residual sum of squares (RSS) (SSR), and available equation below represents:

$SSR={\mathrm{\Σ}}_{i=1}^I{\mathrm{\Σ}}_{j=1}^J{\mathrm{\Σ}}_{k=1}^K{\mathrm{\ϵ}}_{ijk}^2;i=1,\mathrm{....},I;j=1,\mathrm{...},J;k=1,\mathrm{...},K$

In described step d, the concrete calculating process of PARAFAC algorithm is as described below: first, providing fluorescent components predictive value F and model preset value a, b, c value is by three-dimensional response matrix X_ijkAnd a, b preset value judges jointly；Then, a value is together decided on by three-dimensional response matrix X and b, c anticipation value again；Finally, b value is together decided on by three-dimensional response matrix X and a, c anticipation value again, as SSR < 10^-6Time, it is believed that model reaches convergence, provides F, a, b, c value determined。

The invention has the beneficial effects as follows: the invention provides a kind of method that three-dimensional fluorescence spectrum measures pollutant in conjunction with PARAFAC algorithm, based on spectrofluorimetry, method is highly sensitive, favorable reproducibility, easy and simple to handle, analysis cost is cheap, compensate for that traditional chemical detection method process is loaded down with trivial details, it is consuming time to detect, the more high deficiency of testing cost, realize the organic pollution quickly differentiating in the environment such as water body, soil, reference can also be provided for the correlational study of other waste water simultaneously。

Accompanying drawing explanation

In order to be illustrated more clearly that the technical scheme in the embodiment of the present invention, below the accompanying drawing used required during embodiment is described is briefly described:

Fig. 1 is naphthalene luxuriant and rich with fragrance many gradients mixing mark liquid PARAFAC three-dimensional fluorescence spectrum component analysis figure in the present invention。

Detailed description of the invention

In order to be better understood from the present invention, below in conjunction with specific embodiment, invention is described in detail。

A kind of method that three-dimensional fluorescence spectrum measures pollutant in conjunction with PARAFAC algorithm, described method differentiates that step is as follows:

A. the foundation of sample database: initially set up variable concentrations, the different types of pollutant three-dimensional fluorescence spectrum PARAFAC algorithm data-base in particular excitation/emission wavelength ranges, obtain the standard curve of pollutant levels and Characteristic fluorescence intensity simultaneously；

B. sample pre-treatments: be dissolved in deionized water by sample, passes into a certain amount of nitrogen and removes dissolved oxygen then the filtering with microporous membrane with 0.45 μm, remove impurity；

C. the acquisition of three-dimensional fluorescence spectrum data: the sample after removing impurity in step b is taken in 4ml addition cuvette, is placed in three-dimensional fluorescence spectrum analyser, sets parameter, be scanned obtaining the three-dimensional fluorescence spectrum data of sample；

D. the discriminating of pollutant type: the sample three-dimensional fluorescence spectrum data matlab software obtained in step c is loaded in the workbox equipped with PARAFAC algorithm, spectrum is resolved, position according to fluorescence spectral characteristic peak, in conjunction with the three-dimensional fluorescence spectrum data base of pollutant, differentiates the type of pollutant；

E. pollutant levels differentiate: by the pollutant after the determination type that obtains in step d, contrast in the fluorescence intensity of particular excitation/transmitted wave strong point and the intensity of step a standard curve, it is thus achieved that concentration data；

F. Modifying model: the three-dimensional fluorescence spectrum intensity of target contaminant is beyond the detection upper lower limit value of three-dimensional fluorescence spectrum data base, now, the concentration correspondence of target contaminant is diluted or concentration x times, then through step c, d, e, thus corresponding calculating obtains meeting the target pollutant concentration of three-dimensional fluorescence spectrum standard database comparison。

Actually used process: the Organic substance gradient mixed solutions more than 15 parts such as configuration concentration scope naphthalene between 0.10～40.00ng/mL, phenanthrene, anthracene, when specific three dimensional fluorescent scanning: set excitation wavelength between 200-550nm scope, launch wavelength between 250-600nm scope, sweep span scope between 2.5-10nm scope, slit width between 2.5-10nm scope, scanning speed is at 500～2400nm min^-1It is scanned between scope, obtains and set up sample database。Acquired fluorescence data is carried out PARAFAC analysis by offer method in step d, obtains key component and each component X value。Each component and standard substance fluorescence pattern are compared, determine each component materials, it is illustrated in figure 1 gas chromatography (this only provides naphthalene, phenanthrene to illustrate) different gradients for example to mix and mark obtained partial results after solution carries out PARAFAC analysis, analysis result and each sample independent standard solution fluorogram are compared and known judges each component tie substance (C1 is naphthalene, C2 is phenanthrene), and variable concentrations gradient correspondence X is carried out linear regression modeling with concentration value, the detection limit obtained and response parameter are as shown in table 1。

Embodiment 1:

Choose multiple concentration known phenanthrene water body example, pass into nitrogen and remove dissolved oxygen in water body example, by the sample filtering with microporous membrane with 0.45 μm。Post-filtration samples is taken 4mL and adds in four-way cuvette, be placed in and have in the fluorescence spectrum analyzer that excitation-emission synchronizes, arrange excitation wavelength be 200-450nm, to launch wavelength at 280-550nm, sweep span to be 5nm, slit width be 5nm, scanning speed be 1200nm min^-1After, sample is scanned, it is thus achieved that all fluorescence datas。Spectroscopic data is resolved by the Matlab software equipped with PARAFAC algorithmic tool case, select result meets the component of luxuriant and rich with fragrance standard sample characteristic fluorescence spectral peak position, the X data value of its correspondence is derived, is derived in substitution sample database standard curve, it is thus achieved that corresponding concentration data。

Embodiment 2:

Take 5g difference concentration known sample naphthalene contaminated soil sample, with 20mL methanol, it is carried out ultrasonic extraction 30 minutes, after centrifugal 10, reclaim supernatant, repeatedly above step 3 times。By recovered liquid by the C18 extraction column after activating, and carrying out eluting with 10mL normal hexane, after eluting, nitrogen stripping, near dry, be settled to 5mL with normal hexane, take 2mL solution in fluorophotometer sample cell, carry out the mapping of three-dimensional fluorescence spectrum scanning。Instrument parameter is set to: excitation wavelength is 200-450nm, launch wavelength at 280-550nm, sweep span to be 5nm, slit width be 5nm, scanning speed is 1200nm min^-1After。Sample is scanned, it is thus achieved that all fluorescence datas。Spectroscopic data is resolved by the Matlab software equipped with PARAFAC algorithmic tool case, select result meets the component of naphthalene standard sample characteristic fluorescence spectral peak position, the X data value of its correspondence is derived, is derived in substitution sample database standard curve, it is thus achieved that corresponding concentration data。

Table 1 is the detection limit and the response parameter that detect pollution components according to PARAFAC algorithm。

Table 2 is pollutant mark-on testing result in water body in embodiment 1。

Table 3 is pollutant in soil mark-on testing result in embodiment 2。

The detection limit of table 1 pollution components and response parameter

Pollutant (phenanthrene) mark-on testing result in water body in table 2 embodiment 1。

Pollutant in soil (naphthalene) mark-on testing result in table 3 embodiment 2

According to table 1, table 2 and table 3 data this it appears that, the method that a kind of three-dimensional fluorescence spectrum provided by the invention measures pollutant in conjunction with PARAFAC algorithm, by in advance sample database being set up, correlation coefficient is up to more than 0.998, there is accuracy height, highly sensitive, favorable reproducibility, the advantage that easy and simple to handle, analysis cost is cheap, it is achieved that quickly differentiate the organic pollution in the environment such as water body, soil。

The ultimate principle of the present invention, principal character and advantage have more than been shown and described。Skilled person will appreciate that of the industry, above-described embodiment does not limit the present invention in any form, and all employings are equal to the technical scheme that the mode of replacement or equivalent transformation obtains, and all fall within protection scope of the present invention。

Claims (2)

1. the method that a three-dimensional fluorescence spectrum measures pollutant in conjunction with PARAFAC algorithm, it is characterised in that described method differentiates that step is as follows:

A. the foundation of sample database: initially set up variable concentrations, the different types of pollutant three-dimensional fluorescence spectrum PARAFAC algorithm data-base in particular excitation/emission wavelength ranges, obtain the standard curve of pollutant levels and Characteristic fluorescence intensity simultaneously；

B. sample pre-treatments: be dissolved in deionized water by sample, passes into a certain amount of nitrogen and removes dissolved oxygen then the filtering with microporous membrane with 0.45 μm, remove impurity；

C. the acquisition of three-dimensional fluorescence spectrum data: the sample after removing impurity in step b is taken in 4ml addition cuvette, is placed in three-dimensional fluorescence spectrum analyser, sets parameter, be scanned obtaining the three-dimensional fluorescence spectrum data of sample；

D. the discriminating of pollutant type: the sample three-dimensional fluorescence spectrum data matlab software obtained in step c is loaded in the workbox equipped with PARAFAC algorithm, spectrum is resolved, position according to fluorescence spectral characteristic peak, in conjunction with the three-dimensional fluorescence spectrum data base of pollutant, differentiates the type of pollutant；

E. pollutant levels differentiate: by the pollutant after the determination type that obtains in step d, contrast in the fluorescence intensity of particular excitation/transmitted wave strong point and the intensity of step a standard curve, it is thus achieved that concentration data；

F. Modifying model: the three-dimensional fluorescence spectrum intensity of target contaminant is beyond the detection upper lower limit value of three-dimensional fluorescence spectrum data base, now, the concentration correspondence of target contaminant is diluted or concentration x times, then through step c, d, e, thus corresponding calculating obtains meeting the target pollutant concentration of three-dimensional fluorescence spectrum standard database comparison。

2. the method that a kind of three-dimensional fluorescence spectrum according to claim 1 measures pollutant in conjunction with PARAFAC algorithm, it is characterised in that: in described step d, PARAFAC algorithm is: the whole fluorescence datas obtained are built into three-dimensional data matrix X_ijk, three-dimensional data matrix X_ijkIt is decomposed into 3 loading matrixs and residual error number gust, such as drag:

$X_{ijk}={\mathrm{\&Sigma;}}_{f=1}^F{a}_{if}{b}_{jf}{c}_{kf}+{\mathrm{\&epsiv;}}_{ijk};$ I=1 ...., I；J=1 ..., J；K=1 ..., K

X in formula_ijkAt excitation wavelength j and the wavelength k fluorescence intensity data matrix constituted is launched for all samples i；F is the independent fluorescence number of components that system has actual contribution；A_ifFor the relative concentration battle array of f fluorophor in i-th sample；B_jfAnd c_kfTransmitting that respectively f fluorophor is corresponding and excitation spectrum battle array；ε_ijkFor the residual error battle array of model, solving of it is based on the alternately principle of least square and minimize residual sum of squares (RSS) (SSR), and available equation below represents:

$SSR={\mathrm{\&Sigma;}}_{i=1}^I{\mathrm{\&Sigma;}}_{j=1}^J{\mathrm{\&Sigma;}}_{k=1}^K{\mathrm{\&epsiv;}}_{ijk}^2;$ I=1 ...., I；J=1 ..., J；K=1 ..., K

In described step d, the concrete calculating process of PARAFAC algorithm is as described below: first, providing fluorescent components predictive value F and model preset value a, b, c value is by three-dimensional response matrix X_ijkAnd a, b preset value judges jointly；Then, a value is together decided on by three-dimensional response matrix X and b, c anticipation value again；Finally, b value is together decided on by three-dimensional response matrix X and a, c anticipation value again, as SSR < 10^-6Time, it is believed that model reaches convergence, provides F, a, b, c value determined。