Search

CN-122016669-A - Interface characterization method based on synchronous acquisition and decoupling of multi-mode spectrum signals

CN122016669ACN 122016669 ACN122016669 ACN 122016669ACN-122016669-A

Abstract

The invention discloses an interface characterization method based on synchronous acquisition and decoupling of multi-mode spectrum signals, and belongs to the technical field of spectrum detection. According to the method, a single light source is utilized to excite an interface region containing two or more than two different materials, so that the different materials respectively generate spectrum signals with different characteristics, raw data containing various spectrum signals are obtained under the same test condition through a spectrum acquisition device, and the data are decoupled based on a signal separation algorithm to obtain target spectrums respectively corresponding to the different materials. And by combining spectrum acquisition of a plurality of spatial positions, the multi-mode visual analysis of the chemical composition, structural stress and morphology distribution of the interface region can be realized. The invention is suitable for systems with material interfaces such as solid-state batteries, electrochemical catalysis, semiconductor devices and the like, and has the advantages of wide application range, high signal decoupling precision, high detection efficiency and the like.

Inventors

  • ZHANG XING
  • WEI SHENGFU
  • FAN AORAN

Assignees

  • 清华大学

Dates

Publication Date
20260512
Application Date
20260126

Claims (10)

  1. 1. The interface characterization method based on synchronous acquisition and decoupling of the multi-mode spectrum signals is characterized by comprising the following steps of: Applying single light source excitation to interface areas of different materials, and performing spatially resolved spectrum scanning on the interface areas to enable the different materials to generate various spectrum signals with differences; acquiring raw spectral data comprising the plurality of spectral signals; Decoupling the original spectrum data by using a signal separation algorithm to obtain target spectrum signals corresponding to different materials; and extracting characteristic parameters from the target spectrum signal, and quantitatively characterizing the evolution process of the interface region based on the change of the characteristic parameters along with space or time so as to obtain a characterization result.
  2. 2. The method of claim 1, wherein the signal separation algorithm comprises: Carrying out weight distribution on a Raman characteristic peak area and a baseline area in an original spectrum by adopting an asymmetric weight function, and initializing a weight matrix; Introducing smoothness constraint conditions to construct a baseline fitting objective function; based on the previous iteration fitting result, self-adaptively updating a weight matrix, giving lower weight to the Raman peak area, and repeating iteration until the baseline fitting converges; the fitted baseline is separated from the original spectrum to obtain fluorescence signals, and meanwhile, pure Raman characteristic peak signals after the baseline is removed are obtained.
  3. 3. The method of claim 1, wherein after acquiring the raw spectral data comprising the plurality of spectral signals, the method further comprises: Performing wave number axis correction on the original spectrum data to eliminate wave number offset caused by instrument drift or environmental disturbance, and ensuring that spectrum peak positions are aligned under a uniform wave number coordinate system; Carrying out noise evaluation on the spectrum data after wave number correction, and identifying a spectrum region or a space pixel with a signal-to-noise ratio lower than a preset threshold value to obtain a noise evaluation result; And based on the noise evaluation result, adopting a self-adaptive smoothing filtering or wavelet denoising method to optimize the low signal-to-noise ratio region, and outputting the preprocessing spectrum data with improved signal-to-noise ratio and reserved characteristic peak shapes.
  4. 4. The method of claim 1, wherein the different materials produce different spectral signal types under excitation of a single light source, comprising: one material generating a fluorescent signal and the other material generating a Raman signal, or Both materials produce raman signals, but each has non-overlapping or distinguishable characteristic raman peak positions, thereby forming raman signals with differences.
  5. 5. The method of claim 2, wherein decoupling the raw spectral data using a signal separation algorithm to obtain target spectral signals corresponding to different materials comprises: The self-adaptive iteration weighting least square method is adopted to perform fluorescence baseline fitting and separation on the original spectrum, namely, a Raman characteristic peak area and a baseline area are distinguished through an asymmetric weight function, a weight matrix is initialized, smoothness constraint is introduced to construct a baseline fitting objective function, the weight matrix is self-adaptively updated based on the previous iteration result, lower weight is given to the Raman characteristic peak area, iteration is repeated until convergence is achieved, and a separated fluorescence signal and a pure Raman characteristic peak signal are obtained; Analyzing the decoupled multi-mode signals by adopting a principal component analysis algorithm and an independent component analysis algorithm to separate spectral characteristics of different components of an interface to obtain a target spectral signal, wherein the target spectral signal comprises a fluorescent signal corresponding to a fluorescent material and a Raman signal corresponding to a Raman active material.
  6. 6. The method of claim 1, wherein the characteristic parameters include a composition of matter, a concentration of matter, and a local stress, wherein: the substance component is obtained by analyzing the peak position of the decoupled Raman characteristic peak; The substance concentration is obtained by analyzing the peak intensity of the decoupled Raman characteristic peak; The local stress is obtained by analyzing the peak position offset of the decoupled Raman characteristic peak; For a material having no raman characteristic peak but a fluorescence signal, concentration information of the material is obtained by analyzing the degree of spectral baseline elevation.
  7. 7. The method according to claim 6, wherein the substance component is obtained by analyzing peak positions of the decoupled raman characteristic peaks, comprising: a standard Raman characteristic peak position database of a tested interface system is established in advance, and the standard Raman characteristic peak position database contains characteristic peak position information of each component substance; identifying peak position values of characteristic peaks in the decoupled Raman spectrum; and comparing the identified peak position value with a standard database to determine the corresponding substance component.
  8. 8. The method of claim 6, wherein the substance concentration is obtained by analyzing a peak intensity of the decoupled raman characteristic peak, comprising: Pre-establishing a quantitative relation between the Raman characteristic peak intensity and the concentration of each substance component in a tested interface system; extracting the peak intensity value of the corresponding characteristic peak in the decoupled Raman spectrum; and calculating according to the quantitative relation and the extracted peak intensity value to obtain the concentration of the substance.
  9. 9. The method of claim 6, wherein the local stress is obtained by analyzing a peak position shift of the decoupled raman feature peak, comprising: selecting a Raman characteristic peak sensitive to the system stress change of the measured material as a stress probe peak; determining a standard peak position of the stress probe peak in a stress-free state; Measuring the actual peak position of the stress probe peak in the decoupled Raman spectrum; Calculating the offset between the actual peak position and the standard peak position; and determining a local stress value according to a quantitative relation between the pre-established peak position offset and the stress magnitude.
  10. 10. A multi-modal spectral signal simultaneous acquisition and decoupling system for implementing the method of any one of claims 1-9, comprising: a laser light source module configured to emit a single wavelength laser and excite an interface region; The optical path coupling module is in optical path connection with the laser source module and is configured to introduce the single-wavelength laser into the interface area and synchronously collect various mixed spectrum signals generated by the interface area; the spectrum detection module is in optical path connection with the optical path coupling module and is configured to collect the plurality of mixed spectrum signals and convert the mixed spectrum signals into digital signals to be output; The data processing module is in data connection with the spectrum detection module, is internally provided with a preset signal separation algorithm, is configured to receive the digital signals, performs multi-mode signal decoupling processing on the digital signals by utilizing the signal separation algorithm, and outputs target spectrum signals corresponding to different materials and analysis results containing material components, concentrations, local stress and morphology distribution.

Description

Interface characterization method based on synchronous acquisition and decoupling of multi-mode spectrum signals Technical Field The invention relates to the technical field of spectrum detection, in particular to an interface characterization method based on synchronous acquisition and decoupling of multi-mode spectrum signals. Background The microstructure and chemical state of the material interface have a decisive influence on the performance in the fields of energy, electronics, catalysis and the like. For example, in solid state batteries, the interfacial reaction of the electrode with the solid state electrolyte affects ion transport efficiency and cycle life, and in electrochemical catalysis, the interface structure of the catalyst and support directly determines the active site distribution and reaction rate. The method has important significance for real-time and in-situ characterization of the interface in the working state, revealing the interface reaction mechanism and optimizing the material design. Spectral analysis technology is widely used in material research due to its non-destructive, high sensitivity and molecular level information acquisition capability. The raman spectrum provides chemical composition and lattice structure information, while the fluorescence signal can reflect the electronic structure, defect state or morphology characteristics of a specific material. However, in a multi-material interface, spectrum signals of different materials are often coupled with each other, when one of the materials mainly generates a fluorescence signal and the other material generates a raman signal, a fluorescence background can seriously interfere with extraction of the raman signal, when both materials generate the raman signal and characteristic peak intervals are overlapped, peak shapes of the different materials are difficult to distinguish, and in the prior art, signals are distinguished by respectively testing different areas or replacing excitation light sources, so that testing time and operation complexity are increased, and errors caused by environmental changes are easily introduced. In addition, part of methods rely on special hardware or complex light path modification, and are difficult to popularize and use on conventional spectrometer platforms. Thus, there is an urgent need for a method that can simultaneously acquire and effectively separate multiple spectral signals in a single measurement. Disclosure of Invention The present invention aims to solve at least one of the technical problems in the related art to some extent. The invention aims to overcome the defects of difficult information separation, low testing efficiency and high equipment transformation cost caused by multi-material interface spectrum signal coupling in the prior art, and provides an interface characterization method based on synchronous acquisition and decoupling of multi-mode spectrum signals, which can acquire and effectively separate spectrum information of different materials in single measurement and realize in-situ, multi-mode and high-precision characterization of an interface evolution process. Another object of the present invention is to provide a system for synchronously collecting and decoupling multi-mode spectrum signals. In order to achieve the above objective, in one aspect, the present invention provides an interface characterization method based on synchronous acquisition and decoupling of multi-mode spectrum signals, including: Applying single light source excitation to interface areas of different materials, and performing spatially resolved spectrum scanning on the interface areas to enable the different materials to generate various spectrum signals with differences; acquiring raw spectral data comprising the plurality of spectral signals; Decoupling the original spectrum data by using a signal separation algorithm to obtain target spectrum signals corresponding to different materials; and extracting characteristic parameters from the target spectrum signal, and quantitatively characterizing the evolution process of the interface region based on the change of the characteristic parameters along with space or time so as to obtain a characterization result. The interface characterization method based on synchronous acquisition and decoupling of the multi-mode spectrum signals provided by the embodiment of the invention can also have the following additional technical characteristics: In one embodiment of the invention, the fluorescence signal is represented as a baseline elevation and the raman signal is a characteristic peak signal after baseline removal in the original spectrum signal. In one embodiment of the invention, the method is applicable to in situ characterization of solid state batteries, electrochemical catalytic systems, semiconductor devices, or other systems where material interfaces are present. In one embodiment of the invention, the signal separation algorithm co