Plasma spectrum non-normalization calculation method, device and system

Abstract

The invention relates to a plasma spectrum non-normalization calculation method, device and system. The plasma spectrum non-normalization calculation method comprises the following steps: acquiring a spectrum signal of a to-be-detected sample; converting the spectral signal into spectral data; identifying the spectral line broadening of each element according to the spectral data; calculating a self-absorption coefficient of each element according to the spectral line broadening; correcting the spectral line intensity of the optical thin spectral line of each element according to the self-absorption coefficient, and updating and correcting the spectral data; according to the updated and corrected spectral lines of the spectral data, using a Sagoboltzmann plane method to obtain a Sagoboltzmann plane of each element; and calculating the population density of each element according to the SharpBoltzmann plane, and calculating the content of each element according to the population density of each element. The plasma spectrum non-normalization calculation method disclosed by the invention has the advantage of improving the quantitative analysis precision of major elements, trace elements and trace elements.

Description

Technical Field

[0001] The present invention relates to the field of laser spectroscopy analysis, and particularly to a non-normalized calculation method, device, and system for plasma spectra. Background Art

[0002] Laser-induced Breakdown Spectroscopy (LIBS) uses high-energy lasers to excite a plasma with a high temperature electron number density on the surface of a sample, and then analyzes the spectrum emitted by the plasma to obtain qualitative and quantitative analysis of elements in the sample. It is a technique for rapid, non-destructive, and simultaneous detection of multiple elements. Traditional LIBS quantitative analysis methods mainly use normalization models to perform quantitative analysis on sample spectra. The normalization model aims to improve the accuracy and reliability of the model by standardizing and transforming the original data. Therefore, the normalization quantitative analysis method usually depends on normalizing the concentrations of all elements to a certain benchmark, such as the mass percentage of oxides or the element concentration ratio.

[0003] However, during the process of standardizing and transforming the original data by the normalization model, for some elements that are difficult to detect, although they are contained in the sample, in the case where these difficult-to-detect elements cannot be detected, the content of this element will be normalized and assigned to other elements, resulting in some elements being too large. Especially in the detection of trace elements and ultra-trace elements, it will cause a sharp increase in their errors. For example, the presence of nitrogen, water vapor, and oxygen in the ambient gas, especially elements that are difficult to detect, often interfere with the normalization results, making it impossible to obtain the accurate content of trace and ultra-trace elements, resulting in a large deviation in the final result; if there are many elements that are not normalized or the content of the elements that are not normalized is high, it will be catastrophic for the detection of trace and ultra-trace elements. Summary of the Invention

[0004] Based on this, the object of the present invention is to provide a non-normalized calculation method, device, and system for plasma spectra.

[0005] The present invention provides a non - normalization calculation method for plasma spectroscopy, which includes obtaining the spectral signal of a sample to be measured; converting the spectral signal into spectral data; identifying the spectral line broadening of each element according to the spectral data; calculating the self - absorption coefficient of each element according to the spectral line broadening; correcting the spectral line intensity of each element according to the self - absorption coefficient and updating and correcting the spectral data; obtaining the Saha - Boltzmann plane of each element by using the Saha - Boltzmann plane method according to the updated and corrected spectral data; calculating the particle number density of each element according to the Saha - Boltzmann plane, and calculating the content of each element according to the particle number density of each element, specifically including: setting an internal reference element, and selecting a reference sample containing the internal reference element and with a known content; obtaining the intercept of each element in the Saha - Boltzmann plane, and calculating the particle number density of each element according to the intercept; calculating the content of the internal reference element in the sample to be measured; calculating the content of other elements according to the content, particle number density of the internal reference element and the particle number density of other elements, so as to obtain the content of each element.

[0006] Compared with the prior art, the non - normalization calculation method for plasma spectroscopy of the present invention avoids the interference of ambient gas and difficult - to - detect elements in the traditional normalization model by using the non - normalization method of the internal reference element, and improves the quantitative analysis accuracy of major elements while also improving the quantitative analysis accuracy of trace and ultra - trace elements.

[0007] Further, in the step of obtaining the intercept of each element in the Saha - Boltzmann plane and calculating the particle number density of each element according to the intercept, it specifically includes: adding the particle number density of each element in the neutral species state and the particle number density in the singly - ionized species state to obtain the particle number density of each element respectively; the relationship between the particle number density of each element in the neutral species state and the intercept of the element in the Saha - Boltzmann plane is:

[0008]

[0009] The relationship between the particle number density of the singly - ionized species state and the intercept of the element in the Saha - Boltzmann plane is:

[0010]

[0011] In the formula, is the intercept of the Saha - Boltzmann plane in the neutral species state, is the particle number density in the neutral species state, is the intercept of the Saha - Boltzmann diagram in the singly - ionized species state, is the particle number density in the singly - ionized species state, and All are partition functions related to temperature; the expression of the partition function US(T) is as follows:

[0012]

[0013] In the formula, gi is the energy level degeneracy, Ei is the lower energy level, kB is the Boltzmann constant, and T is the plasma temperature.

[0014] Further, in the step of calculating the content of the internal reference element in the sample to be measured, it specifically includes: according to the sum of the particle number densities of the selected internal reference element in the neutral state and the singly ionized species state, combined with the content and particle number density of the internal reference element in the reference sample, and calculating the content of the internal reference element in the sample according to the element content calculation formula. The element content calculation formula is as follows:

[0015]

[0016] In the formula, C I is the content of the internal reference element in the sample to be measured, m I is the relative atomic mass of the internal reference element in the sample to be measured, FN I is the particle number density of the internal reference element in the sample to be measured, m II is the relative atomic mass of the internal reference element in the reference sample, FN II is the particle number density of the internal reference element in the reference sample, C II is the content of the internal reference element in the reference sample.

[0017] Further, in the step of calculating the content of each other element according to the content, particle number density of the internal reference element and the particle number density of each other element, so as to obtain the content of each element, the content of the other element is obtained through the particle number density, relative atomic mass of the other element and the particle number density, relative atomic mass of the internal reference element. Its formula expression is as follows:

[0018]

[0019] In the formula, mS is the relative atomic mass of the other element, Cs represents the content of the other element, FNs is the particle number density of the other element, mi is the relative atomic mass of the internal reference element, Ci represents the content of the internal reference element, and FNi is the particle number density of the internal reference element.

[0020] Further, in the process of identifying the spectral line broadening of each element according to the spectral data, it specifically includes:

[0021] Calculate the maximum decomposition level according to the length of the spectral data. The calculation formula for the maximum decomposition level is as follows:

[0022] J = log2(N),

[0023] where J is the length of the spectral data, and N is the length of the spectral data;

[0024] Decompose the spectral data into multiple decomposed spectral data according to all possible decomposition levels within the maximum decomposition level; each of the decomposed spectral data is decomposed based on a different decomposition level; each decomposed spectral data is divided into a low-frequency approximation component and a high-frequency detail component based on the decomposition level, where the low-frequency approximation component is decomposed into the low-frequency approximation component of the high layer and the low-frequency approximation component of the remaining layers;

[0025] For each of the decomposed spectral data, retain the low-frequency approximation component of the high layer, set to zero or correct the low-frequency approximation component of the remaining layers, and perform threshold processing on the high-frequency detail component to obtain the processed low-frequency approximation component and high-frequency detail component;

[0026] Reconstruct the spectral data based on the processed low-frequency approximation component and high-frequency detail component in each of the decomposed spectral data under different decomposition levels to obtain corrected spectral data with background removal under different decomposition levels;

[0027] Calculate the spectral similarity Q1 and the background removal integrity Q2 of each corrected spectral data respectively, and add the spectral similarity Q1 and the background removal integrity Q2 of each corrected spectral data as the comprehensive scoring coefficient of the corrected spectral data at this decomposition level; the calculation formula for the spectral similarity Q1 is as follows:

[0028]

[0029] where y is the original signal, y′ is the corrected signal, cov is the covariance, and σ is the standard deviation;

[0030] The calculation formula for the background removal integrity Q2 is as follows:

[0031]

[0032] where max(y′) is the maximum value of the corrected signal, and σ y′ is the standard deviation of the local minimum;

[0033] Retain the corrected spectral data with the highest comprehensive scoring coefficient as the obtained spectral data after background removal.

[0034] Further, in the process of calculating the self-absorption coefficient of each element according to the spectral line broadening, it specifically includes performing a second derivative process on the spectral data to obtain the second derivative of the wavelength of the spectral data with respect to the spectral intensity; detecting the number of peaks and the center position of each peak in the spectral data according to the second derivative of the spectral data, performing peak splitting fitting on the spectral data, calculating the full width at half maximum after peak splitting, and obtaining the spectral line width.

[0035] Further, after obtaining the Saha-Boltzmann plane of each element by using the Saha-Boltzmann plane method according to the spectral line of the updated and corrected spectral data, the following steps are further included: obtaining the column density Boltzmann plane by using the column density Boltzmann plane method for the spectral line of the spectral data that has not been updated and corrected and is converted from the spectral signal; obtaining the accuracy of self-absorption correction by comparing the slopes of the Saha-Boltzmann plane and the column density Boltzmann plane.

[0036] Further, it further includes: when the difference between the slopes of the Saha-Boltzmann plane and the column density Boltzmann plane is greater than a preset threshold, sequentially using the updated and corrected spectral data to perform the steps of identifying the spectral line broadening of each element according to the spectral data, calculating the self-absorption coefficient of each element according to the spectral line broadening, correcting the spectral line intensity of each element according to the self-absorption coefficient, updating and correcting the spectral data, and obtaining the Saha-Boltzmann plane of each element by using the Saha-Boltzmann plane method according to the updated and corrected spectral data; obtaining the Saha-Boltzmann plane of each element by using the Saha-Boltzmann plane method, and rechecking the accuracy of the self-absorption correction obtained this time, continuously performing iterative operations until the plasma temperature error calculated by the Saha-Boltzmann plane and the column density Boltzmann plane is less than the preset threshold, then stopping the iteration, and outputting the updated and corrected spectral data obtained in the last iterative operation.

[0037] Based on the same inventive concept, the present invention also provides a non-normalized calculation device for plasma spectroscopy, which includes a spectral signal acquisition module for acquiring the spectral signal of a sample to be measured; a spectral data conversion module for converting the spectral signal into spectral data; a spectral line broadening identification module for identifying the spectral line broadening of each element according to the spectral data; a self-absorption coefficient calculation module for calculating the self-absorption coefficient of each element according to the spectral line broadening; a self-absorption correction module for correcting the spectral line intensity of each element according to the self-absorption coefficient and updating and correcting the spectral data; a Boltzmann plane calculation module for obtaining the Saha-Boltzmann plane of each element by using the Saha-Boltzmann plane method according to the updated and corrected spectral data; and an element content calculation module for calculating the particle number density of each element according to the Saha-Boltzmann plane and calculating the content of each element according to the particle number density of each element.

[0038] Based on the same inventive concept, the present invention further provides a non-normalized calculation system for plasma spectra, which includes a laser module for generating a laser to break through the surface of a sample to form plasma; a sensor module for collecting the optical signal emitted by the plasma on the surface of the sample; and a data processor for performing the non-normalized calculation method for plasma spectra according to any one of claims 1-8 to obtain the content of each element of the sample.

[0039] For better understanding and implementation, the present invention will be described in detail below with reference to the accompanying drawings. Description of the Drawings

[0040] Figure 1 It is a schematic diagram of the modules of the non-normalized calculation system for plasma spectra of the present invention;

[0041] Figure 2 It is a schematic diagram of the modules of the non-normalized calculation device for plasma spectra of the present invention;

[0042] Figure 3 It is a schematic flowchart of the non-normalized calculation method for plasma spectra of the present invention;

[0043] Figure 4 It is a flowchart of step S2 of the non-normalized calculation method for plasma spectra of the present invention;

[0044] Figure 5 It is a schematic diagram of spectral data;

[0045] Figure 6 It is a schematic diagram of the second derivative of the spectral data;

[0046] Figure 7 It is a Saha-Boltzmann plane drawn based on the spectral line before self-absorption correction;

[0047] Figure 8 It is a Saha-Boltzmann plane drawn based on the spectral line after self-absorption correction;

[0048] Figure 9 It is a comparison diagram of the column density Boltzmann plane and the Saha-Boltzmann plane of the first soil sample. Detailed Embodiments

[0049] In view of the situation that the existing plasma spectrum normalization model is likely to cause a sharp increase in errors of trace elements and ultratrace elements, the inventor designed a calculation method for the non-normalized plasma spectrum model. This calculation method can calculate the content of each element in the sample without performing a normalization operation on the original data, improving the accuracy of calculating the element content, especially the accuracy of the content of trace elements and ultratrace elements. In addition, the inventor also corrects the spectral self-reversal caused by the self-absorption phenomenon generated inside the sample, and calculates the accuracy of the corrected result.

[0050] Please refer to Figure 1 , Figure 1 which is a schematic diagram of the modules of the non-normalized plasma spectrum calculation system of the present invention. A non-normalized plasma spectrum calculation system of the present invention is used to detect the content of each element in a sample, and includes a laser module 1, a sensor module 2, and a data processor 3. The laser module 1 generates a laser and focuses the laser on the surface of the sample to ablate the surface of the sample to generate a plasma. The sensor module 2 collects the optical signal of the plasma generated by the laser ablation of the surface of the sample by the laser module 1 and converts the optical signal into a spectral signal based on an electrical signal. The data processor 3, which is the non-normalized plasma spectrum calculation device of the present invention, executes the non-normalized plasma spectrum calculation method of the present invention according to the spectral signal collected by the sensor module 2 to calculate the content of each element in the sample.

[0051] Please refer to Figures 2-3 , Figure 2 which is a schematic diagram of the modules of the non-normalized plasma spectrum calculation device of the present invention, Figure 3 which is a schematic diagram of the process of the non-normalized plasma spectrum calculation method of the present invention. The non-normalized plasma spectrum calculation device of the present invention is the data processor 3, which includes: a spectral signal acquisition module M0, a spectral data conversion module M1, a spectral line broadening identification module M2, a self-absorption coefficient calculation module M3, a self-absorption correction module M4, a Boltzmann plane calculation module M5, and an element content calculation module M6.

[0052] The spectral signal acquisition module M0 is used to execute step S0: acquire the spectral signal of the sample to be measured.

[0053] The spectral data conversion module M1 is used to execute step S1: convert the spectral signal into spectral data. The spectral data is an optically thick spectral line, whose abscissa is the wavelength and the ordinate is the spectral intensity.

[0054] Please refer to Figure 4 , Figure 4This is a flowchart of step S2 of the non-normalized calculation method for plasma spectrum of the present invention. The spectral line broadening identification module M2 is used to execute step S2: identify the spectral line broadening of each element according to the spectral data. This step S2 specifically includes:

[0055] S21: Calculate the maximum decomposition level according to the length of the spectral data. The calculation formula for the maximum decomposition level is:

[0056] J = log2(N),

[0057] where J is the length of the spectral data and N is the length of the spectral data.

[0058] S22: Decompose the spectral data into multiple decomposed spectral data according to all possible decomposition levels within the maximum decomposition level; each of the decomposed spectral data is decomposed based on a different decomposition level; each decomposed spectral data is divided into a low-frequency approximation component and a high-frequency detail component based on the decomposition level, where the low-frequency approximation component is decomposed into the low-frequency approximation component of the high layer and the low-frequency approximation component of the remaining layers. The reference wavelet is the basis function of wavelet transform and determines the characteristics of spectral decomposition; in this embodiment, the db7 wavelet in Daubechies wavelets is selected as the reference wavelet to decompose the spectral data. The length of the spectral data is the wavelength span in the spectral data, that is, in the decomposed spectral data, each decomposition layer corresponds to the spectral data of a specific frequency band. The low-frequency approximation component usually represents the low-frequency background trend, and the background noise mainly exists in this component; each layer of the low-frequency approximation component is a further subdivision of the upper-level low-frequency signal, and the low-frequency approximation component is divided into the highest layer and the remaining layers. The high-frequency detail component mainly retains the details of the spectral data. All possible decomposition levels within the maximum decomposition level refer to the decomposition levels that are less than the maximum decomposition level and can decompose the spectral data.

[0059] S23: For each of the decomposed spectral data, retain the low-frequency approximation component of the high layer, set to zero or correct the low-frequency approximation component of the remaining layers, and perform threshold processing on the high-frequency detail component to obtain the processed low-frequency approximation component and high-frequency detail component.

[0060] S24: Reconstruct the spectral data according to the processed low-frequency approximation component and high-frequency detail component in each of the decomposed spectral data under different decomposition levels to obtain the corrected spectral data with background removal under different decomposition levels.

[0061] S25: Calculate the spectral similarity Q1 and the background removal integrity Q2 of each corrected spectral data respectively and add them up to obtain the comprehensive scoring coefficient of the corrected spectral data at each decomposition level; the calculation formula of the spectral similarity Q1 is:

[0062]

[0063] In the formula, y is the original signal, y′ is the corrected signal, cov is the covariance, and σ is the standard deviation;

[0064] The calculation formula of the background removal integrity Q2 is:

[0065]

[0066] In the formula, max(y′) is the maximum value of the corrected signal, and σ y′ is the standard deviation of the local minimum. The spectral similarity Q1 measures the similarity degree of two or more spectra in dimensions such as shape, intensity, or characteristic peaks, and it is calculated using the Pearson correlation coefficient. The background removal integrity Q2 is a comprehensive evaluation index of the retention degree of the target signal and the elimination effect of background noise after removing the background interference in the spectrum through algorithms or technical means.

[0067] S26: Retain the corrected spectral data with the highest comprehensive scoring coefficient as the obtained spectral data after background removal. The obtained spectral data after background removal can effectively remove most of the background on the basis of maintaining the spectral peak shape compared with the original spectral data. At the same time, due to its adaptive characteristics, this method can be used for any length and any type of laser-induced breakdown spectrum without a large number of adjustments and optimizations.

[0068] Please refer to Figures 5-6 , Figure 5 for the schematic diagram of spectral data, Figure 6 and Figure 5 for the schematic diagram of the second derivative of the spectral data. The spectral line broadening identification module M2 is used to execute step S2: identify the spectral line broadening of each element according to the spectral data after background removal. Perform a second derivative process on the spectral data after background removal to obtain the second derivative of the wavelength of the spectral data with respect to the spectral loudness; detect the number of peaks and the center position of each peak in the spectral data according to the second derivative of the spectral data, and perform peak splitting fitting on the spectral data to calculate the full width at half maximum after peak splitting to obtain the spectral line width. Compare the Figure 5 and Figure 6It can be seen from the spectral data and the second derivative of the spectral data that the troughs of the second derivative of the spectral data, i.e., the positions marked by dots, correspond one-to-one with the peaks of the spectral data, i.e., the positions marked by x's. Furthermore, in the case of overlapping peaks and continuous peaks, the recognition of the number of peaks of the spectral data and the central position of each peak is ensured, so as to obtain the accurate spectral line widths of each element.

[0069] The self-absorption coefficient calculation module M3 is used to execute step S3: calculate the self-absorption coefficient of each element according to the spectral line broadening. The self-absorption coefficient is the ratio of the actually detected spectral line broadening to the theoretical broadening. The theoretical broadening Δλ0 is approximated by Stark broadening. The formula for the self-absorption coefficient is expressed as:

[0070]

[0071] In the formula, Δλ is the actually detected spectral line broadening, Δλ0 is the theoretical broadening under ideal conditions, and SA is the self-absorption coefficient. The calculation formula for Stark broadening is expressed as:

[0072]

[0073] In the formula, ω s is the Stark broadening coefficient, N ref is the relative electron number density, and N e is the electron number density. The calculation method for the electron number density N e is as follows:

[0074]

[0075] In the formula, Δλ 1 / 2 is the full width at half maximum of the actually measured H α line, which is fitted by the Lorentz function, and α 1 / 2 is the half height width of the reduced Stark line shape. Since in the research of LIBS, the Gaussian broadening of the spectral line broadening can be ignored compared with the Lorentz broadening. Therefore, the Lorentz line width is obtained by the approximate formula of subtracting the Gaussian broadening from the actually detected spectral line broadening. This Lorentz approximation formula is expressed as:

[0076]

[0077] In the formula, Δλ meas is the actually measured spectral line broadening, which is obtained by fitting with the Voigt function; Δλ G = λ / R, where λ is the wavelength and R is the resolution of the spectral signal acquisition module M0.

[0078] The self-absorption correction module M4 is used to execute step S4: according to the self-absorption coefficient, correct the spectral line intensities of each element, and update and correct the spectral data. To correct the spectral line intensities of each element, the true spectral line intensity is divided by the self-absorption coefficient to obtain the corrected spectral line intensity. Subsequently, the corrected spectral line intensity and the theoretical broadening calculated in step S3 are updated and corrected into the spectral data obtained in step S1 to obtain an updated and corrected spectrogram, which contains the spectral lines of the corrected spectral line broadening and spectral line intensity of each element. Self-absorption correction can bring the obtained optically thick spectral lines to the state of optically thin spectral lines; the optically thin spectral lines are spectral lines not affected by the self-absorption effect, but the number of such spectral lines is small and often needs to be corrected by other spectral lines.

[0079] The Boltzmann plane calculation module M5 executes step S5: according to the updated and corrected spectral data, use the Saha-Boltzmann plane method to obtain the Saha-Boltzmann plane of each element. The updated and corrected spectral data is the spectral data obtained in step S4. The Saha-Boltzmann plane method describes the relationship between the particle number densities of neutral and singly ionized state particles of the same element based on the Saha-Eggert equation, which is applicable to the condition of local thermal equilibrium (LTE), and its formula is expressed as:

[0080]

[0081] In the formula, N II and N I are the number densities of atomic state particles and singly ionized state particles respectively, E ion is the first ionization energy, m e is the mass of an electron, h is Planck's constant, k B is the Boltzmann constant, U I (T) and U II (T) are the partition functions of each state.

[0082] The content calculation module M6 of each element is used to execute step S6: calculate the particle number density of each element according to the Saha-Boltzmann plane, and calculate the content of each element according to the particle number density of each element. Step S6 includes:

[0083] S61: Set an internal reference element, and select a reference sample containing this internal reference element and with a known content. Select an element with relatively stable concentration in the matrix of the sample to be measured as the internal reference element. The reference sample is a standard sample containing the internal reference element, or other samples for which the content of the internal reference element has been measured; meanwhile, the particle number density of the internal reference element in this reference sample is also known.

[0084] S62: Obtain the intercepts of each element in the Saha-Boltzmann plane, and calculate the particle number density of each element according to the intercepts. Based on the Saha-Boltzmann plane obtained in step S above, add the particle number density of each element in the neutral species state to the particle number density in the singly ionized species state to obtain the particle number density of each element respectively. Among them, the relationship between the particle number density of each element in the neutral species and the intercept of the element in the Saha-Boltzmann plane is:

[0085]

[0086] The relationship between the particle number density in the singly ionized species state and the intercept of the element in the Saha-Boltzmann plane is:

[0087]

[0088] In the formula, is the intercept of the Saha-Boltzmann diagram in the neutral species state, is the particle number density in the neutral species state, is the intercept of the Saha-Boltzmann diagram in the singly ionized species state, is the particle number density in the singly ionized species state, and are both partition functions related to temperature; the expression of the partition function U S (T) is:

[0089]

[0090] In the formula, g i is the energy level degeneracy, E i is the lower energy level, k B is the Boltzmann constant, and T is the plasma temperature.

[0091] S63: Calculate the content of the internal reference element in the sample to be tested. According to the sum of the particle number densities of the selected internal reference element in the neutral state and the singly ionized species state, combined with the content and particle number density of the internal reference element in the reference sample, calculate the content of the internal reference element in this sample according to the element content calculation formula. The element content calculation formula is:

[0092]

[0093] In the formula, C I is the content of the internal reference element in the sample to be tested, m I is the relative atomic mass of the internal reference element in the sample to be tested, FN Iis the particle number density of the internal reference element in the sample to be measured, m II is the relative atomic mass of the internal reference element in the reference sample, FN II is the particle number density of the internal reference element in the reference sample, C II is the content of the internal reference element in the reference sample. Since the internal reference elements in the reference sample and the sample to be measured are the same element and have equal relative atomic masses, that is, when calculating, m I = m II； At this time, through the content and particle number density of the internal reference element in the reference sample, the content of the internal reference element in the sample to be measured can be calculated.

[0094] S64: Calculate the contents of other elements based on the content, particle number density of the internal reference element and the particle number densities of other elements, so as to obtain the contents of all elements. According to the content of the internal reference element, compare the particle number densities of the internal reference element and other elements, and calculate the contents of other elements one by one. The formula for calculating the content of other elements is expressed as

[0095]

[0096] In the formula, m S is the relative atomic mass of other elements, Cs represents the content of other elements, FN S is the particle number density of the other element, m i is the relative atomic mass of the internal reference element, C i represents the content of the internal reference element, FN i is the particle number density of the internal reference element.

[0097] Further, in order to evaluate the accuracy of the correction of the spectral line intensity in step S4, after step S4, the following steps are also included: obtaining a column density Boltzmann plane for optically thick spectral lines in the spectral data that has not been updated and corrected according to the spectral signal by using the column density Boltzmann plane method; obtaining the accuracy of the self-absorption correction by comparing the slopes of the Saha Boltzmann plane and the column density Boltzmann plane. Among them, the slope of each spectral line in the Saha Boltzmann plane represents the first plasma temperature, and the slope of each spectral line in the column density Boltzmann plane represents the second plasma temperature; comparing the first plasma temperature and the second plasma temperature, that is, comparing the slopes in the Saha Boltzmann plane and the column density Boltzmann plane, and evaluating the accuracy of the self-absorption correction through the proximity of the slopes. When the slopes in the Saha Boltzmann plane and the column density Boltzmann plane are close, that is, the first plasma temperature and the second plasma temperature are close, the accuracy of the self-absorption correction is high. The column density Boltzmann diagram method is an improvement of the Saha Boltzmann plane method, which can directly use the temperature evaluation under optically thick plasma conditions. It describes the plasma by directly using the column density of ground-state or low-energy-state particles, without self-absorption correction, avoiding the limitation of the optically thin assumption. The optically thin assumption is a common physical and astronomical simplified model used to describe the cut-off where light can escape freely. The expression of the column density Boltzmann plane method is:

[0098]

[0099] where is the column density of ions at a lower energy level; n I l is the column density of neutral particles. The calculation formula for the column density is:

[0100]

[0101] In the formula, k(λ0 ) l is the optical depth, which depends on the self-absorption degree of the spectral line, and its relationship with the self-absorption coefficient (SA) is:

[0102]

[0103] Further, when the plasma temperatures calculated through the Saha-Boltzmann plane and the column density Boltzmann plane differ by more than a preset threshold, steps S3 - S6 and the calibration accuracy assessment are re-performed using the updated and corrected spectral data, and the iterative calculation is continuously carried out until the plasma temperature error calculated through the Saha-Boltzmann plane and the column density Boltzmann plane is less than the preset threshold, at which point the iteration stops and the updated and corrected spectral data obtained in the last iterative calculation is output. In this embodiment, the preset threshold is 10%; when the temperature error calculated from the Saha-Boltzmann diagram and the column density Boltzmann diagram reaches more than 10%, it indicates that the accuracy of the self-absorption correction is relatively low. At this time, steps S6 - S8 of the self-absorption correction are repeated using the updated and corrected spectral data until the error between the two is less than 10%, and then the iteration stops. This method of continuously iterating to improve the accuracy of self-absorption correction, based on the characteristics of the column density Boltzmann diagram that can fully utilize the self-absorption effect, can not only more accurately eliminate the influence of self-absorption, but also introduce a feedback mechanism for automatically judging the convergence of the correction, thereby enhancing the robustness of the overall system.

[0104] The technical effects of the present invention are illustrated by experimental data as follows:

[0105] The plasma spectral non-normalization calculation method is used to calculate the system for detecting the first soil sample. The element contents in the first soil sample are shown in the following table, where the element contents are expressed in concentration:

[0106]

[0107] Please refer to Figures 7-8 , Figure 7 which is the Saha-Boltzmann plane drawn based on the spectral line before self-absorption correction, Figure 8 and

[0108] which is the Saha-Boltzmann plane drawn based on the spectral line after self-absorption correction. During the detection of the first soil sample, the parallelism of the spectral lines of each element in the Saha-Boltzmann plane diagram drawn according to the spectral line before self-absorption correction is relatively low; while in the Saha-Boltzmann plane diagram drawn based on the spectral line after self-absorption correction, the parallelism of the spectral lines of each element is significantly improved. Therefore, the self-absorption correction using this method can effectively correct the self-absorption effect generated during the detection process. Figure 9 , Figure 9It is a comparison chart of the column density Boltzmann plane and the Saha Boltzmann plane of the first soil sample. The column density Boltzmann plane is drawn for the Mg element and Ca element of the first soil sample, and the Saha Boltzmann plane is drawn for the spectral lines of the Si, Al, Ca, Fe, and Mg elements in the updated and corrected spectral data. The slopes of the elements in the column density Boltzmann plane and the Saha Boltzmann plane are close, indicating that the derived plasma temperatures are almost the same, that is, the accuracy of the self-absorption correction is high.

[0109] The second soil sample is detected by the system using the non-normalized calculation method of the plasma spectrum. The contents of its various elements are shown in the following table, where the element contents are expressed in concentration:

[0110]

[0111] Using the non-normalized calculation method of the plasma spectrum of the present invention to detect the contents of Si element and Ti element therein, wherein the first soil sample is used as a reference sample and the Si element is used as an internal reference element. It is detected that the content of Si element is 29.095%, and the relative error (REs) is 5.493%; the content of Ti element is 0.313%, and the relative error is 11.664%.

[0112] In addition, for the third soil sample and the fourth soil sample, the non-normalized calculation method of the plasma spectrum of the present invention is used to detect the contents of Si element and Ti element therein, and the first soil sample is also used as a reference sample and the Si element is used as an internal reference element.

[0113] The content of Si element in the third soil is 28.709%, and the content of Ti element is 0.61%. The detection result using the non-normalized calculation method of the plasma spectrum of the present invention is that the content of Si element is 26.969%, and the relative error is 6.061%; the content of Ti element is 0.673%, and the relative error is 10.274%.

[0114] The content of Si element in the third soil is 32.396%, and the content of Ti element is 0.394%. The detection result using the non-normalized calculation method of the plasma spectrum of the present invention is that the content of Si element is 34.442%, and the relative error is 6.316%; the content of Ti element is 0.409%, and the relative error is 4.318%.

[0115] Based on the above results of detecting the second to third soil samples using the non-normalized calculation method of the plasma spectrum of the present invention, it can be seen that the detection result of the non-normalized calculation method of the plasma spectrum of the present invention has a small error, good stability, can have good detection accuracy for different samples, and has small fluctuations.

[0116] Comparative example

[0117] In the literature "Quantitative deteermination of element concentraions in industrial oxide materials by laser - industusced breakdown spectroscopy", in the detection of the major element Si and the trace element Ti in soil samples using the normalization model method, the relative error of the Si element detection is 33%, and the relative error of the Ti element detection is 46%.

[0118] Compared with the comparative example, the element content detected by using the non - normalization calculation method of the plasma spectrum of the present invention is significantly more accurate. Whether for the No. 2 soil sample, No. 3 soil sample or No. 4 soil sample, the maximum relative error of the detection result is 6.316% for the Si element and 11.664% for the Ti element, which are much smaller than the relative error of 33% for the Si element and 46% for the Ti element in the comparative example. At the same time, since the soil samples in the comparative example and the No. 2 soil sample, No. 3 soil sample or No. 4 soil sample all have the Si element as the major element and the Ti element as the trace element, it can be seen that the non - normalization calculation method of the plasma spectrum of the present invention is superior to the existing detection method using the normalization model both in detecting major elements and trace elements.

[0119] Compared with the prior art, the non - normalization calculation method of the plasma spectrum of the present invention, by using the non - normalization method of internal reference elements, avoids the interference of environmental gases and difficult - to - detect elements in the traditional normalization model, and improves the quantitative analysis accuracy of trace and ultra - trace elements. At the same time, it corrects the element self - absorption phenomenon during the detection process, and evaluates the accuracy of the self - absorption correction by calculating the plasma temperature, thereby obtaining the accuracy of the detected element content and improving the reliability of the detection method.

[0120] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments and are not intended to limit the embodiments of the present application. The singular forms "a", "the" and "said" used in the embodiments of the present application and the claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that unless otherwise stated, "a plurality of" means two or more; the terms "first", "second", "third", etc. are only used for distinction and not for describing a specific order or sequence, nor can they be understood as indicating or implying relative importance. The term "and / or" used herein means any or all possible combinations of one or more of the associated listed items. When the above description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. In the description of the present application, for those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0121] The above embodiments only represent several implementation manners of the present invention. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and the present invention also intends to include these modifications and improvements.

Claims

1. A plasma spectrum non-normalization calculation method, characterized in that: Includes steps: Acquiring a spectral signal of a sample to be tested; converting the spectral signal into spectral data; Identifying the line broadening of each element based on the spectral data; Calculating the self-absorption coefficient of each element according to the spectral line broadening; According to the self-absorption coefficient, the spectral line intensity of each element is corrected, and the spectral data is updated and corrected; According to the updated and corrected spectral data, the Saha-Boltzmann plane of each element is obtained using the Saha-Boltzmann plane method; The particle number density of each element is calculated according to the Saha-Boltzmann plane, and the content of each element is calculated according to the particle number density of each element, specifically including: Setting an internal reference element and selecting a reference sample containing the internal reference element and having a known content; Obtaining the intercept of each element in the Saha-Boltzmann plane, and calculating the particle number density of each element according to the intercept; Calculating the content of the internal reference element in the sample to be tested; The content of other elements is calculated according to the content of the internal reference element, the particle number density and the particle number density of other elements, so as to obtain the content of each element.

2. The plasma spectrum non-normalization calculation method according to claim 1, characterized in that: The step of obtaining the intercept of each element in the Saha-Boltzmann plane and calculating the particle number density of each element according to the intercept specifically includes: adding the particle number density of each element in the neutral species state and the particle number density in the single ionized species state to obtain the particle number density of each element; the relationship between the particle number density of each element in the neutral species state and the intercept of the element in the Saha-Boltzmann plane is: The relationship between the particle number density in the single ionized species state and the intercept of the element in the Saha-Boltzmann plane is: In the formula, is the intercept of the Saha-Boltzmann plane in the neutral species state, is the particle number density in the neutral species state, is the intercept of the Saha-Boltzmann diagram in the single ionized species state, is the particle number density in the single ionized species state, and are all temperature-dependent partition functions; the partition function U S The expression of (T) is: In the formula, g i is the energy level degeneracy, E i is the lower level energy, k B is the Boltzmann constant, and T is the plasma temperature.

3. The plasma spectrum non-normalization calculation method according to claim 2, characterized in that: The step of calculating the content of the internal reference element in the sample to be tested specifically includes: according to the sum of the particle number density of the selected internal reference element in the neutral state and the single ionized species state, combined with the content and particle number density of the internal reference element in the reference sample, the content of the internal reference element in the sample is calculated according to the element content calculation formula, and the element content calculation formula is: In the formula, C I is the content of the internal reference element in the sample to be tested, m I is the relative atomic mass of the internal reference element in the sample to be tested, FN I is the particle number density of the internal reference element in the sample to be tested, m II is the relative atomic mass of the internal reference element in the reference sample, FN II is the particle number density of the internal reference element in the reference sample, C II is the content of the internal reference element in the reference sample.

4. The plasma spectrum non-normalization calculation method according to claim 3, characterized in that: In the step of calculating the content of each element according to the content, particle number density and particle number density of the internal reference element, so as to obtain the content of each element, the content of other elements is obtained by the particle number density and relative atomic mass of the other elements and the particle number density and relative atomic mass of the internal reference element, and the formula is expressed as follows: In the formula, m S is the relative atomic mass of the other element, Cs represents the content of the other element, FNs is the particle number density of the other element, m i is the relative atomic mass of the internal reference element, Ci represents the content of the internal reference element, and FNi is the particle number density of the internal reference element.

5. The plasma spectrum non-normalization calculation method according to claim 1, characterized in that: The process of identifying the spectral line broadening of each element according to the spectral data specifically includes: The maximum number of decomposition layers is calculated according to the length of the spectral data. The calculation formula of the maximum number of decomposition layers is: J = log2(N), Wherein, J is the length of the spectral data, and N is the length of the spectral data; According to all possible decomposition levels within the maximum decomposition level, the spectral data is decomposed into a plurality of decomposed spectral data; each of the decomposed spectral data is decomposed based on a different number of decomposition levels; each of the decomposed spectral data is divided into a low-frequency approximate component and a high-frequency detail component based on the number of decomposition levels, wherein the low-frequency approximate component is decomposed into the low-frequency approximate component of the high-level layer and the low-frequency approximate component of the remaining layers; For each of the decomposed spectral data, retain the low-frequency approximate components of the high-level layers, set to zero or correct the low-frequency approximate components of the remaining layers, and perform threshold processing on the high-frequency detail components to obtain the processed low-frequency approximate components and high-frequency detail components; Reconstructing the spectrum data according to the processed low-frequency approximate components and high-frequency detail components in each of the decomposed spectrum data at different decomposition levels to obtain corrected spectrum data with background removed at different decomposition levels; The spectral similarity Q1 and the background removal completeness Q2 of each corrected spectral data are calculated respectively, and the spectral similarity Q1 and the background removal completeness Q2 of each corrected spectral data are added as the comprehensive scoring coefficient of the corrected spectral data under the decomposition level; the calculation formula of the spectral similarity Q1 is: In the formula, y is the original signal, y′ is the corrected signal, cov is the covariance, and σ is the standard deviation; The calculation formula of the background removal completeness Q2 is: Where max(y′) is the maximum value of the corrected signal, σ y′ is the standard deviation of the local minimum; The corrected spectral data with the highest comprehensive scoring coefficient is retained as the obtained spectral data after background removal.

6. The plasma spectrum non-normalization calculation method according to claim 1, characterized in that: The process of calculating the self-absorption coefficient of each element according to the spectral line broadening specifically includes: performing second-order derivative processing on the spectral data to obtain the second-order derivative of the wavelength of the spectral data with respect to the spectral intensity; detecting the number of peaks in the spectral data and the center position of each peak according to the second-order derivative of the spectral data, performing peak fitting on the spectral data, calculating the half-height full width after peak fitting, and obtaining the spectral line width.

7. The plasma spectrum non-normalization calculation method according to claim 1, characterized in that: After obtaining the Saha Boltzmann plane of each element according to the spectral lines of the updated and corrected spectral data using the Saha Boltzmann plane method, the method also includes the steps of: obtaining a column density Boltzmann plane for the spectral lines of the spectral data that have not been updated and corrected according to the spectral signal conversion using the column density Boltzmann plane method; and obtaining the accuracy of the self-absorption correction by comparing the slopes of the Saha Boltzmann plane and the column density Boltzmann plane.

8. The plasma spectrum non-normalization calculation method according to claim 7, characterized in that: It also includes: when the difference between the slopes of the Saha-Boltzmann plane and the column density Boltzmann plane is greater than a preset threshold, using the updated and corrected spectral data to re-perform the steps in sequence: identifying the spectral line broadening of each element according to the spectral data, calculating the self-absorption coefficient of each element according to the spectral line broadening, correcting the spectral line intensity of each element according to the self-absorption coefficient, updating and correcting the spectral data, and obtaining the Saha-Boltzmann plane of each element using the Saha-Boltzmann plane method according to the updated and corrected spectral data; The Saha-Boltzmann plane method is used to obtain the Saha-Boltzmann plane of each element, and the accuracy of the self-absorption correction obtained this time is re-checked, and the iterative operation is continuously performed until the plasma temperature error calculated by the Saha-Boltzmann plane and the column density Boltzmann plane is less than the preset threshold value. The iteration is stopped, and the updated and corrected spectral data obtained in the last iterative operation is output.

9. A plasma spectrum non-normalization calculation device, characterized in that: include: A spectral signal acquisition module is used to acquire the spectral signal of the sample to be tested; A spectral data conversion module, used for converting the spectral signal into spectral data; A spectral line broadening identification module, used to identify the spectral line broadening of each element according to the spectral data; A self-absorption coefficient calculation module, used to calculate the self-absorption coefficient of each element according to the spectral line broadening; A self-absorption correction module, used to correct the spectral line intensity of each element according to the self-absorption coefficient, and update and correct the spectral data; A Boltzmann plane calculation module, used for obtaining the Saha-Boltzmann plane of each element using the Saha-Boltzmann plane method according to the updated and corrected spectral data; The element content calculation module is used to calculate the particle number density of each element according to the Saha-Boltzmann plane, and calculate the content of each element according to the particle number density of each element.

10. A plasma spectrum non-normalization calculation system, characterized in that: include: A laser module is used to generate laser light, so that the laser light penetrates the surface of the sample to form plasma; A sensor module, used for collecting light signals emitted by the plasma on the surface of the sample; A data processor is used to execute the plasma spectrum non-normalization calculation method described in any one of claims 1 to 8 according to the optical signal to obtain the content of each element in the sample.

Images