Overlapped spectral line separation method based on MPT (Microwave Plasma Torch) spectral data

Abstract

The invention discloses an overlapped spectral line separation method based on MPT (Microwave Plasma Torch) spectral data. The method comprises the following steps: selecting to-be-analyzed overlappedspectral data, and estimating feature data of to-be-separated spectral lines in the overlapped spectral data, wherein the feature data comprises a maximum peak I, a peak wavelength lambda and a half-peak breadth sigma; fitting an effective analysis area corresponding to the maximum peak I; constructing an objective function based on an MPT spectral mathematical model, and performing optimal matching on the feature data of to-be-separated spectral lines in the effective analysis area by utilizing the objective function so as to obtain the optimal separated spectral line; and subtracting the spectral data of the optimal separated spectral line from the overlapped spectral data, thereby obtaining the next to-be-analyzed overlapped spectral data. Therefore, when sample analysis is performed by an MPT spectrometer, qualitative and quantitative analysis can be performed by directly using the data in an overlapped peak area, so that the analytical performance of the MPT spectrometer and thedetection ability and detection precision of the MPT spectrometer on complicated mixed samples are greatly improved.

Description

technical field [0001] The invention relates to the technical field of atomic emission spectroscopic analysis, in particular to a method for separating overlapping spectral lines based on MPT spectral data. Background technique [0002] The MPT spectrometer is composed of a light source system and a spectral data acquisition and analysis system. The light source system is used to excite the characteristic spectrum of the elements in the sample to be tested; the spectrum acquisition and analysis system is used to collect the characteristic spectrum information of the elements excited by the light source system, and then use the characteristic spectrum information for qualitative or quantitative analysis of elements in a sample. The separation of each component of the sample to be analyzed at different wavelength positions on the spectrogram is completed by the spectroscopic system in the spectral data acquisition and analysis system. However, due to the limited resolution cap...

AI Technical Summary

Problems solved by technology

The separation of each component of the sample to be analyzed at different wavelength positions on the spectrogram is completed by the spectroscopic system in the spectral data acquisition and analysis system. However, due to the limited resolution capability of the hardware equipment, many element spectral lines with very close wavelengths are separated. The spectral peaks cannot be completely separated, and in the collected characteristic spectral data, it appears as a combined peak form formed by the overlapping of multiple spectral peaks, which is not conducive to the subsequent quantitative analysis of samples

[0003] Therefore, under normal circumstances, since the overlapping peak data is composed of the spectral peaks of multiple element spectral lines, it cannot be used directly. Therefore, when using the MPT spectrometer for sample analysis, try to avoid using the spectral lines in the overlapping peak-shaped area. , which makes many times, in order to avoid overlapping peaks, give up the selection of the spectral line with strong emission intensity of the element to be measured, and choose the spectral line with weak emission intensity of the element, which reduces the detection precision and accuracy of the instrument for the sample.

Especially when analyzing complex mixed samples, it is necessary to carry out complex sample pretreatment first. Even after pretreatment, it is often due to the strong optional emission intensity and few or no peak-overlapping spectral lines, resulting in the instrument’s sensitivity to the sample. The detection precision and accuracy are not enough, even the samples cannot be quantitatively detected

Images