Dynamic calibration method for echelle spectrometer in laser-induced breakdown spectroscopy

Abstract

The present invention belongs to the technical field of elemental analysis, and more particularly, relates to a dynamic calibration method for echelle spectrometer in laser-induced breakdown spectroscopy, comprising: S1: collecting a standard light source by using an echelle spectrometer; S2: in combination with a calibration function, calculating a pixel position coordinate ({circumflex over (x)},) corresponding to a spectral wavelength ; S3: performing dynamic searching and filtering near the pixel position coordinate ({circumflex over (x)},) to obtain a set D of all pixel position coordinates, and adjusting all original intensity values in the set D to obtain intensity values F(I.sub.x,y), and S4: calculating a spectral line intensity value after dynamic calibration by summing the adjusted intensity values F(I.sub.x,y), thereby completing dynamic calibration of the result of the echelle spectrometer. The method in the present invention can overcome the shortcoming, i.e., the existing echelle spectrometer is only calibrated before measurement without solving the spectral line drift during use, increasing the absolute intensity of the wavelength and reducing the detection limit of the quantitative analysis, as well as improving the precision of the quantitative analysis of an element to be analyzed.

Claims

1. A dynamic calibration method for echelle spectrometer in laser-induced breakdown spectroscopy, comprising: S1: by using an echelle spectrometer, collecting a spectrogram of a standard calibration light source, and getting spectral wavelengths in different orders according to the spectrogram; S2: based on information about the spectral wavelengths in different orders obtained in the step S1, and in combination with a corresponding calibration function (x,y), calculating a pixel position coordinate ({circumflex over (x)},) corresponding to a spectral wavelength of a certain analysis element, where x and y respectively represent original pixel horizontal and vertical coordinates of an analysis element on the echelle spectrometer, and {circumflex over (x)} and respectively represent pixel horizontal and vertical coordinates corresponding to a spectral wavelength of a certain analysis element on the echelle spectrometer; S3: by using the echelle spectrometer, collecting laser-induced plasma spectra corresponding to the above analysis element in a sample to be tested, performing dynamic searching and filtering near the pixel position coordinate ({circumflex over (x)},) obtained in the step S2 to obtain a set D of all selected pixel position coordinates, and adjusting all original intensity values in the set D to obtain adjusted intensity values expressed by an adjustment function F(I.sub.x,y), where I.sub.x,y represents a spectral line intensity value at the pixel position coordinate (x,y), and x and y in the pixel position coordinate (x,y) in the set D respectively have lower limit values of in and n, and respectively have upper limit values of m and n; and S4: by using the adjusted intensity values F(I.sub.x,y) obtained in the step S3, calculating a spectral line intensity value at the spectral wavelength after dynamic calibration according to the following formula: .sub.=.sub.x,y=m,n.sup.mnF(I.sub.x,y), which is used as the spectral line intensity value at the wavelength of the analysis element, so that the dynamic calibration for the given spectral line of the echelle spectrometer is completed, where I.sub.x,y represents a spectral line intensity value at the pixel position coordinate (x,y) on the echelle spectrometer, the function F is an adjustment function, .sub. represents a spectral line intensity value corresponding to the spectral wavelength after calibration and is a sum function.

2. The dynamic calibration method of the echelle spectrometer of claim 1, wherein a selection range of the set D of pixel position coordinates in the step S3 is determined by the following steps: by taking a spectral intensity {circumflex over (q)} as a judgment intensity, performing dynamic searching near the pixel position coordinate ({circumflex over (x)},), and if a synthetical intensity q at a certain pixel position coordinate is smaller than the judgment intensity {circumflex over (q)}, not adding the pixel position coordinate into the set D, otherwise, adding the pixel position coordinate into the set D.

3. The dynamic calibration method of the echelle spectrometer of claim 2, wherein the calculation functions of the synthetical intensity q and the judgment intensity {circumflex over (q)} are as follow:
q=[exp(z)1].Math.q.sub.0
{circumflex over (q)}=q.sub.0/2
z=k.Math.(x{circumflex over (x)}).sup.2+l.Math.(y).sup.2, where z represents the distance between the pixel position coordinates (x,y) and ({circumflex over (x)},), q.sub.0 represents the original intensity at the pixel position coordinate ({circumflex over (x)},), exp is an exponential function, and k and l represent weight coefficients which are determined by users in actual use.

4. The dynamic calibration method of the echelle spectrometer of claim 3, wherein the dynamic searching near the pixel position coordinate ({circumflex over (x)},) in the step S3 is performed in a spiral movement manner.

5. The dynamic calibration method of the echelle spectrometer of claim 4, wherein the adjustment function F in the step 3 is as follow:
F=I.sub.x,y+|I.sub.x,yI.sub.{circumflex over (x)},|.Math.[1exp(z)].Math.v, where z represents the distance between the pixel position coordinates (x,y) and ({circumflex over (x)},), I.sub.{circumflex over (x)}, represents a spectral line intensity value at the pixel position coordinate ({circumflex over (x)},), I.sub.x,y represents a spectral line intensity value at the pixel position coordinate (x,y), exp is an exponential function, and v represents an adjustment coefficient which is determined by users in actual use.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a flowchart of a dynamic calibration method for echelle spectrometer in LIBS according to the present invention;

(2) FIGS. 2(a) and 2(b) shows intensity drafting of LIBS on the detector of the echelle spectrometer;

(3) FIGS. 3(a) and 3(b) are comparison diagrams of results with and without an intensity adjustment function for LIBS according to the present invention; and

(4) FIGS. 4(a) and 4(b) show fitted curves for manganese in a test sample adopting the conventional method and the method according to the present invention, respectively.

DETAILED DESCRIPTION OF THE EMBODIMENTS

(5) For clear understanding of the objectives, features and advantages of the present invention, detailed description of the present invention will be given below in conjunction with accompanying drawings and specific embodiments. It should be noted that the embodiments described herein are only meant to explain the present invention, and not to limit the range of the present invention.

(6) FIG. 1 is a flowchart of a dynamic calibration method for echelle spectrometer in LIBS according to the present invention, and as shown in FIG. 1, the method specifically comprising the following steps:

(7) S1: by using an echelle spectrometer, collecting a spectrogram of a standard calibration light source, and getting spectral wavelengths in different orders according to the spectrogram;

(8) S2: based on information about the spectral wavelengths in different orders obtained in the step S1, and in combination with a corresponding calibration function (x,y), calculating a pixel position coordinate ({circumflex over (x)},) corresponding to a spectral wavelength of a certain analysis element,

(9) where x and y respectively represent original pixel horizontal and vertical coordinates of an analysis element on the echelle spectrometer, and {circumflex over (x)} and respectively represent pixel horizontal and vertical coordinates corresponding to a spectral wavelength of a certain analysis element on the echelle spectrometer;

(10) S3: by using the echelle spectrometer, collecting laser-induced plasma spectra corresponding to the above analysis element in a sample to be tested, performing dynamic searching and filtering near the pixel position coordinate ({circumflex over (x)},) obtained in the step S2 to obtain a set D of all selected pixel position coordinates, and adjusting all original intensity values in the set D to obtain adjusted intensity values expressed by an adjustment function F(I.sub.x,y),

(11) where I.sub.x,y represents a spectral line intensity value at the pixel position coordinate (x,y), and x and y in the pixel position coordinate (x,y) in the set D respectively have lower limit values of m and n, and respectively have upper limit values of m and n; and

(12) S4: by using the adjusted intensity values F(I.sub.x,y) obtained in the step S3, calculating a spectral intensity at the spectral wavelength after dynamic calibration according to the following formula: .sub.=.sub.x,y=m,n.sup.mnF(I.sub.x,y), which is used as the spectral intensity at the wavelength of the analysis element, so that the dynamic calibration for the given spectral line of the echelle spectrometer is completed,

(13) where I.sub.x,y represents a spectral line intensity value at the pixel position coordinate (x,y) on the echelle spectrometer, the function F is an adjustment function, .sub. represents a spectral line intensity value corresponding to the spectral wavelength after calibration and is a sum function.

(14) In a preferred embodiment of the present invention, a selection range of the set D of pixel position coordinates in the step S3 is determined by the following steps: by taking a spectral intensity {circumflex over (q)} as a judgment intensity, performing dynamic searching near the pixel position coordinate ({circumflex over (x)},), and if a synthetical intensity q at a certain pixel position coordinate is smaller than the judgment intensity {circumflex over (q)}, not adding the pixel position coordinate into the set D, otherwise, adding the pixel position coordinate into the set D. Many comparative experiments showed that after dynamic searching, useless pixels can be excluded to reduce the background intensity of the wavelength and then reduce the intensity fluctuation of the characteristic wavelength of the element, thereby effectively improving the analytical precision of quantitative calculation.

(15) In another preferred embodiment of the present invention, the calculation functions of the synthetical intensity q and the judgment intensity {circumflex over (q)} are as follow:
q=[exp(z)1].Math.q.sub.0
{circumflex over (q)}q.sub.0/2
z=k.Math.(x{circumflex over (x)}).sup.2+l.Math.(y).sup.2,

(16) where z represents the distance between the pixel position coordinates (x,y) and ({circumflex over (x)},), q.sub.0 represents the original intensity at the pixel position coordinate ({circumflex over (x)},) exp is an exponential function, and k and l represent weight coefficients which are determined by users in actual use.

(17) In another preferred embodiment of the present invention, the dynamic searching near the pixel position coordinate ({circumflex over (x)},) in the step S3 is performed in a spiral movement manner. Many comparative experiments showed that by designing an appropriate synthetical intensity calculation function and selecting an appropriate judgment intensity, useless pixels can be excluded to reduce the background intensity of the wavelength and then reduce the intensity fluctuation of the characteristic wavelength of the element, thereby effectively improving the analytical precision of quantitative calculation.

(18) In another preferred embodiment of the present invention, the adjustment function F in the step 3 is as follow:
F=I.sub.x,y+|I.sub.x,yI.sub.{circumflex over (x)},|.Math.[1exp(z)].Math.v,

(19) where z represents the distance between the pixel position coordinates (x,y) and ({circumflex over (x)},), I.sub.{circumflex over (x)}, represents a spectral line intensity value at the pixel position coordinate ({circumflex over (x)},), I.sub.x,y represents a spectral line intensity value at the pixel position coordinate (x,y), exp is an exponential function, and v represents an adjustment coefficient which is determined by users in actual use.

(20) For better explanation of the present invention, a specific embodiment is given below.

(21) For the spectrographs of LIBS with an echelle spectrometer, the spectrographs tend to fluctuate within a small range due to the fluctuation characteristics of the plasma, as shown in FIG. 2. This fluctuating phenomenon is generated due to the spatiotemporal instability of the plasma generated by laser-matter interaction, and does not appear in other application fields of the echelle spectrometer. Therefore, this is also a prerequisite and a substantial problem to be solved in the method of the present invention.

(22) In a first step, an echelle spectrometer is calibrated by a stabilized light source (a mercury lamp for spectrometer calibration), and linear fitting is performed on the calibrated spectral lines in all spectral orders to be processed so as to obtain a wavelength calibration function in a certain spectral order:
w=a.Math.x.sup.2+b.Math.y.sup.2+c.Math.x.Math.y+d.Math.x+e.Math.y+,

(23) where a, b, c, d, e and are coefficients of the quadratic fitting curve, x and y are pixel position coordinate values on the echelle spectrometer detector. It should be noted here that the specific function used for wavelength calibration is not limited to the quadratic fitting function here, and any fitting function that converts a pixel position coordinate into a wavelength is included.

(24) In a second step, for a certain wavelength (e.g., 403.2 nm) of an element (e.g., manganese) to be analyzed in a sample, its coordinate values on the detector are calculated. Since there may be multiple pixel position coordinates that satisfy the calculation result within a given error range, intensity values at the multiple pixel position coordinates are arranged and then a pixel position coordinate corresponding to the maximum intensity value is selected and denoted by {circumflex over (x)} and .

(25) It should be noted here that the determination of this pixel position coordinate is not limited to the use of the maximum intensity value of a single pixel as a criterion, and 22, 33 or multiple other combinations of pixel unit sizes may be selected. In addition, the intensity can also be calculated by weighted summation of the respective intensities of multiple pixel units. The essence of this step is to select the core imaging position of the plasma corresponding to this wavelength on the spectrometer detector, and any calculation process for this purpose can be used in this step.

(26) In a third step, by taking the pixel position coordinate {circumflex over (x)} and as an initial position, a synthetical intensity q at the pixel position coordinate x and y near the pixel position coordinate {circumflex over (x)} and is calculated, and the calculation functions of the synthetical intensities q are as follow:
q=[exp(z)1].Math.q.sub.0
z=k.Math.(x{circumflex over (x)}).sup.2+l.Math.(y).sup.2,

(27) where z represents the distance between the pixel position coordinates (x,y) and ({circumflex over (x)},), q.sub.0 represents the original intensity at the pixel position coordinate ({circumflex over (x)},), exp is an exponential function, and k and l represent weight coefficients which are determined by users in actual use.

(28) The judgment intensity {circumflex over (q)} can be calculated by the following the formula:
{circumflex over (q)}=q.sub.0/2

(29) It should be noted here that the core of this step is to select all pixels in the appropriate range by taking the pixel position coordinate {circumflex over (x)} and obtained in the second step as the central point, and any calculation process for this purpose can be used in this step.

(30) In a fourth step, the original intensity values at the pixel position coordinates obtained in the third step are adjusted, and the adjusted intensity values are summed to obtain a result value, which is used as the intensity value of the wavelength (403.2 nm) of the element (manganese) in the sample to be tested, and participates in subsequent quantitative or qualitative calculations.

(31) The adjustment function F is as follow:
F=I.sub.x,y+|I.sub.x,yI.sub.{circumflex over (x)},|.Math.[1exp(z)].Math.v.

(32) It should be noted here that that the core of this step is to dynamically adjust the original pixel intensity values selected in the third step, and any calculation process for this purpose can be used in this step.

(33) Quantitative analysis of elements Mn, Cr and V in microalloyed steels is performed by the above method:

(34) The experimental device is a standard laser-induced breakdown spectrometer. The experiment is carried out by an echelle spectrometer (Andor Technology, Mechelle5000) in an air atmosphere. A Nd:YAG pulsed laser (Brilliant B, wavelength: 532 nm) is used with a laser repetition frequency of 10 Hz and an average pulse energy of 70 mJ. Via a mirror and a convex lens, the laser is focused onto the surface of a sample which is placed on a three-dimensional platform for planar circular radiation motion. In the experiment, standard samples of microalloyed steel (No.: GSB03-2453-2008) are selected and the experiment is repeated 40 times for each sample.

(35) A wavelength of 403.2 nm is selected for the analysis element of Mn. In this case, this method reduces the detection limit by about 12.5% compared to the traditional quantitative analysis results, and the linear fitting correlation coefficient of the linear calibration curve is increased by about 1%. FIG. 4 show a comparison result of a fitted curve adopting the conventional method and a fitted curve adopting this method for the element Mn, in which FIG. 4(a) shows a fitted curve adopting the conventional method and FIG. 4(b) shows a fitted curve adopting this method.

(36) A wavelength of 427.4 nm is selected for the analysis element of Cr. In this case, this method reduces the detection limit by about 10.2% compared to the traditional quantitative analysis results, and the linear fitting correlation coefficient of the linear calibration curve is increased by about 2%.

(37) A wavelength of 440.8 nm is selected for the analysis element of V. In this case, this method reduces the detection limit by about 13.6% compared to the traditional quantitative analysis results, and the linear fitting correlation coefficient of the linear calibration curve is increased by about 5%.

(38) While particular embodiments of the present invention have been shown and described, it will be obvious to those skilled in the art that changes and modifications may be made without departing from the spirit and range of the present invention.