Noise processing method for surface enhanced in-situ electrochemical infrared data

Abstract

The invention relates to the technical field of electrochemical experiments, in particular to a noise processing method for surface-enhanced in-situ electrochemical infrared data. The method comprises the following steps: introducing nitrogen with the purity of 99.99% into an infrared spectrometer to remove water vapor, and automatically and continuously collecting spectral information under different voltages at an interval of 0.05 V in a range of-0.8-0.8 V until the double-peak intensity of carbon dioxide is reduced to basically disappear from 2240-2400 cm <-1 >; converting background data A and background data B collected for the last two times in the spectral information to obtain an infrared spectrogram C of water vapor; performing data maximum and minimum normalization on the infrared spectrogram C to obtain normalized water vapor peak data D; deducting a carbon dioxide peak from the water vapor peak data D in a range of 2200-3000 cm <-1 > to obtain a water vapor infrared spectrum E; the infrared spectrum data absorbance F to be deducted and the water vapor spectrum intensity numerical value matrix E are imported into MATLAB2023B software, after deduction calculation is conducted, the deducted infrared spectrum data are output, and compared with traditional manual deduction, the method is more efficient and accurate.

Description

Technical Field

[0001] The invention relates to the technical field of electrochemical experiments, and in particular to a noise processing method for surface enhanced in-situ electrochemical infrared data. Background Art

[0002] Surface-enhanced electrochemical in-situ infrared technology has important scientific significance and application value. It can monitor molecular reactions on the electrode surface in real time and provide direct evidence for studying molecular reaction mechanisms and kinetic processes. Through the surface enhancement effect, this technology significantly improves the detection ability of low-concentration molecules and weak signals, effectively distinguishes small changes in electrochemical reactions, and provides molecular structure information to deeply understand the surface chemical and physical properties of materials. With the deepening of research in fields such as catalysis and energy storage, in-situ infrared spectroscopy technology has received increasing attention due to its potential in monitoring complex reactions, studying reaction mechanisms, and discovering intermediates.

[0003] However, the large amount of infrared spectral data generated in practical applications is often interfered by water vapor, resulting in weak signals and difficult to analyze. Generally, expensive vacuum infrared and long-term nitrogen purge are used to reduce the interference of water vapor and carbon dioxide. These methods are time-consuming, labor-intensive and extremely costly. Therefore, it is particularly important to develop efficient, accurate and economical data processing methods to reduce these interference effects.

[0004] The traditional water vapor subtraction method mainly relies on manual adjustment of the difference spectrum method, which is calculated by instrument software (such as Thermo Fisher's OMNIC). Researchers need to try many times (usually 10-100 times) to obtain the appropriate subtraction effect. This method is not only time-consuming and labor-intensive, but also prone to errors due to human judgment, which significantly reduces the reliability of the analysis results.

[0005] In view of this, the present invention provides a noise processing method for surface enhanced in-situ electrochemical infrared data. Summary of the invention

[0006] In order to solve the problem that infrared spectral data is often interfered by water vapor, the traditional subtraction method mainly relies on manual adjustment, which has the disadvantages of low efficiency, large error and weak signal. Therefore, the present invention provides a noise processing method for surface enhanced in situ electrochemical infrared data, and the specific steps of the method are as follows: S1. Introduce 99.99% pure nitrogen into the infrared spectrometer to remove water vapor and carbon dioxide, and automatically and continuously collect spectral information at different voltages within the range of -0.8-0.8V at intervals of 0.05V until the carbon dioxide double peak intensity is between 2240-2400cm -1 Become smaller until almost disappearing; S2. converting the background data A and B collected at the last two times in the above spectral information to obtain an infrared spectrum C of water vapor; S3. Performing maximum and minimum normalization on the infrared spectrum C to obtain normalized water vapor peak data D; S4. The above water vapor peak data D is at 2200-3000cm -1 The carbon dioxide peak is subtracted from the range to obtain the water vapor infrared spectrum E; S5. Import the infrared spectrum data absorbance F to be deducted and the water vapor spectrum intensity numerical matrix E into the MATLAB2023B software, perform the deduction calculation, and then output the infrared spectrum data after deduction.

[0007] Preferably, in S1, the frequency of automatically and continuously collecting spectral information under different voltages is once every 120 seconds.

[0008] Preferably, in S2, the conversion relationship between the background data A, B and the infrared spectrum is: C=log(A / B).

[0009] Preferably, in S3, the maximum and minimum normalized range of the infrared spectrum C data is 0-0.005, and the specific normalization formula is: Among them, c is the original data; Min(c) is the minimum value in the original data; Max(c) is the maximum value in the original data; D is the normalized water vapor peak data.

[0010] Preferably, in S4, the water vapor peak data D is between 2200-3000cm -1 The formula for deducting the carbon dioxide peak from the range is: Where y(3000) is the absorbance value at wave number 3000; y(2200) is the absorbance value at wave number 2200; x(3000) is the abscissa at wave number 3000; x(2200) is the abscissa at wave number 2200; yi is the absorbance value after processing from wave number 2200 to 3000; D2=[y(2200),yi,……,y(3000)]; D = {D1, D2, D3}; E={D1,D2,D3}; D1 is 1200-2200cm in data D -1 Datasets; D2 is 2200-3000cm in data D -1 Datasets; D3 is 3000-4000cm in data D -1 Dataset.

[0011] Preferably, in S5, the deduction calculation formula is: X = F + κ * E; The spectral data X is [X1, X2, X3…Xm]; κ is the reduction factor.

[0012] Preferably, the deduction coefficient κ is in the range of [-100100] with an interval of 0.01.

[0013] Preferably, whether the subtraction calculation is completed is determined by the size of the standard deviation σ of the spectral data X, and the calculation method of the standard deviation σ is as follows: Y=[X(2)X(1),X(3)X(2),…,X(m)X(m1)]; σ represents the standard deviation, yi represents each element in the data set Y, μ represents the mean value of the data set, m-1 represents the number of elements in the dataset.

[0014] Preferably, 3500-4000cm -1 The standard deviation σ within the band range is used as the standard to judge whether the total intensity value of the spectrum after subtraction reaches the minimum value. When the total value reaches the minimum, the subtraction is considered to be completed.

[0015] Compared with the prior art, the present invention has the following advantages: The present invention is different from the traditional manual debugging method. It uses the infrared signal of water vapor in the infrared spectrum signal mainly in the range of 3500-4000cm -1 The minimum standard deviation of the smoothest waveform in the band is obtained through continuous iterative calculations, and the infrared spectrum with the smallest total spectral intensity value is selected. This method avoids the tedious process of manually adjusting the coefficients, shortens the deduction time to a few seconds, and greatly improves the efficiency of data processing. It is particularly suitable for processing large quantities of infrared spectral data, and the algorithm can determine whether the deduction is completed by itself, ensuring the accuracy of spectral deduction. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a schematic diagram of the gradual weakening of the double peak intensity of water vapor and carbon dioxide during the data collection stage of the present invention; Figure 2 is the infrared spectrum of water vapor in the test example of the present invention; Figure 3 This is the infrared spectrum of water vapor after deducting the background drift in the test example of the present invention; Figure 4 The infrared spectrum of the test example of the present invention contains a large number of interference peaks; Figure 5 This is the infrared spectrum after artificially deducting water vapor in the test example of the present invention; Figure 6 This is the infrared spectrum obtained after using the subtraction algorithm of the present invention in the test example of the present invention. DETAILED DESCRIPTION

[0017] In order to deepen the understanding of the present invention, the present invention will be further described in detail below in conjunction with examples. The examples are only used to explain the present invention and do not constitute a limitation on the protection scope of the present invention.

[0018] First, for data acquisition, the in-situ electrochemical device was installed, the optical path was adjusted, and nitrogen with a purity of 99.99% was introduced into the instrument to remove water vapor and carbon dioxide. Then, the Seris automatic acquisition program in the Omnic software of the Thermo Fisher infrared spectrometer IS50 was used to automatically and continuously acquire spectra, collecting background data every 120 seconds until 2240-2400 cm -1 The intensity of the carbon dioxide double peaks in the range decreases to almost disappear. Figure 2 shown.

[0019] Since the infrared background is collected in a single beam form, and the ordinary infrared spectrum is presented in the form of absorbance. The absorbance needs to be obtained by logarithmic transformation of the single beam data. Therefore, in order to obtain the spectrum, the infrared spectrum data needs to be divided by the background, that is, A / B is calculated, and then the absorbance graph is obtained through logarithmic transformation. The specific conversion formula is C=log(A / B). In order to remove the influence and interference of the double peak intensity of carbon dioxide on the spectrum, A and B are the last two background data collected in the above data collection stage.

[0020] Then, the above data C is normalized to the maximum and minimum values ​​between 0 and 0.005. The formula is as follows: Among them, C is the original data; Min(c) is the minimum value in the original data; Max(c) is the maximum value in the original data; D is the normalized water vapor peak data.

[0021] Then, the water vapor peak data D at 2200-3000cm -1The carbon dioxide peak is deducted from the range, and the interference of carbon dioxide is deducted to obtain the water vapor infrared spectrum E. The specific formula is: Where y(3000) is the absorbance value at wave number 3000; y(2200) is the absorbance value at wave number 2200; x(3000) is the abscissa at wave number 3000; x(2200) is the abscissa at wave number 2200; yi is the absorbance value after processing from wave number 2200 to 3000; D2=[y(2200),yi,……,y(3000)]; D = {D1, D2, D3}; E={D1,D2,D3}; D1 is 1200-2200cm in data D -1 Datasets; D2 is 2200-3000cm in data D -1 Datasets; D3 is 3000-4000cm in data D -1 Dataset.

[0022] Finally, the infrared spectrum data absorbance F and the water vapor spectrum intensity numerical matrix E to be deducted are imported into MATLAB2023B, and the deduction calculation is performed according to the following formula: X = F + κ * E; The spectral data X is [X1, X2, X3…Xm]; κ is the reduction factor.

[0023] Among them, because the original data signal strength is about 0.01-0.1, and the water vapor signal strength is 0.005, it can be seen that the original data signal strength is 2-20 times that of the water vapor signal. And the signal has positive and negative, so multiplying the water vapor spectrum intensity value by [-100100] is a suitable interval, that is, the value range of the subtraction coefficient κ is [-100100]. At the same time, in order to avoid too many data operations, affecting the subtraction efficiency, or too few data operations to obtain the best water vapor subtraction effect, the selection interval of the subtraction coefficient κ is set to 0.01.

[0024] In this way, after the infrared spectrum data absorbance F is calculated by subtraction, the spectrum data set X will be obtained. However, how to find the optimal subtraction data in the spectrum data after the above-mentioned large amount of subtraction becomes a new problem. Therefore, in addition to the above-mentioned subtraction calculation, the calculation of the standard deviation σ needs to be introduced.

[0025] The specific calculation method of standard deviation σ is as follows: Y=[X(2)X(1),X(3)X(2),…,X(m)X(m1)]; σ represents the standard deviation, yi represents each element in the data set Y, μ represents the mean value of the data set, m-1 represents the number of elements in the dataset.

[0026] Since the infrared spectrum signal is mainly concentrated in 1200-3500cm -1 band, so in 3500-4000cm -1 In the range, except for the free hydroxyl peak of water vapor, generally organic matter will not have a signal in this area. The infrared signal of water vapor is mainly between 1300-2000cm -1 and 3500-4000cm -1 So, in the 3500-4000cm -1 Within this band, if the spectral curve is smooth enough, it means that the water vapor peak is subtracted more thoroughly, and within this band, since there are no spectral signals of other organic matter, there is no need to worry about false subtraction. It can ensure that the interference of water vapor signal is subtracted while retaining the infrared signal of electrochemical reaction, thus ensuring the accuracy of spectral signal subtraction.

[0027] Therefore, when we are at 3500-4000cm -1 In this band, the computer is used to automatically select the spectral data X corresponding to the smallest standard deviation σ. Then the spectral data X is the one with the smallest total spectral intensity value after deduction, that is, the spectral data after deduction that we need.

[0028] Experimental example: In the experiment of platinum-plated catalyst on the surface of gold-coated silicon crystal, 500 ml / min of nitrogen gas with a purity of 99.99% was first introduced into the instrument to purge water vapor and carbon dioxide. A single-beam spectrum was collected every two minutes until the carbon dioxide peak basically disappeared.

[0029] The last two background data A and B are processed using the following formula: log(A / B) to obtain the water vapor infrared spectrum, such as Figure 3 Then normalization and carbon dioxide peak subtraction are performed to obtain the water vapor infrared spectrum, as shown in Figure 4 shown.

[0030] Methanol was added into the system, and a surface enhanced cyclic voltammetry electrochemical experiment was conducted in the voltage range of -0.8-0.8 volts and with a sampling interval of 0.05 v.

[0031] In the experiment, 32 groups of peaks containing a large number of interference peaks (such as water vapor) were obtained. Figure 5 The infrared spectrum data shown. These large amounts of water vapor peaks will seriously interfere with the spectrum analysis, and the traditional manual analysis method is used to deduct the above 32 spectra one by one, which takes about 2 hours.

[0032] By using the noise processing method provided by the present invention, the researchers imported experimental data and Figure 2 The algorithm successfully removed the interference peaks after multiple iterations and output an infrared spectrum with clear signals and less interference in about 1 second. Figure 6 .

[0033] The above experimental results show that the noise processing method of the present invention greatly reduces the time of manual adjustment and avoids the errors caused by subjective judgment. When processing large-scale data, the algorithm's automatic processing ability is particularly outstanding, which significantly improves the efficiency of spectral analysis.

Claims

1. A noise processing method for surface enhanced in situ electrochemical infrared data, characterized in that: The specific steps of this method are as follows: S1. Introduce 99.99% pure nitrogen into the infrared spectrometer to remove water vapor and carbon dioxide, and automatically and continuously collect spectral information at different voltages within the range of -0.8-0.8V at intervals of 0.05V until the carbon dioxide double peak intensity is between 2240-2400cm -1 Become smaller until almost disappearing; S2. converting the background data A and B collected at the last two times in the above spectral information to obtain an infrared spectrum C of water vapor; S3. Performing maximum and minimum normalization on the infrared spectrum C to obtain normalized water vapor peak data D; S4. The above water vapor peak data D is at 2200-3000cm -1 The carbon dioxide peak is subtracted from the range to obtain the water vapor infrared spectrum E; S5. Import the infrared spectrum data absorbance F to be deducted and the water vapor spectrum intensity numerical matrix E into the MATLAB2023B software, perform the deduction calculation, and then output the infrared spectrum data after deduction.

2. The noise processing method for surface enhanced in-situ electrochemical infrared data according to claim 1, characterized in that: In S1, the frequency of automatically and continuously collecting spectral information under different voltages is once every 120 seconds.

3. The noise processing method for surface enhanced in-situ electrochemical infrared data according to claim 1, characterized in that: In S2, the conversion relationship between the background data A, B and the infrared spectrum is: C=log(A / B).

4. The noise processing method for surface enhanced in-situ electrochemical infrared data according to claim 1, characterized in that: In S3, the maximum and minimum normalized range of the infrared spectrum C data is 0-0.005, and the specific normalization formula is: ； Among them, c is the original data; Min(c) is the minimum value in the original data; Max(c) is the maximum value in the original data; D is the normalized water vapor peak data.

5. The noise processing method for surface enhanced in-situ electrochemical infrared data according to claim 1, characterized in that: In S4, the water vapor peak data D is at 2200-3000cm -1 The formula for deducting the carbon dioxide peak from the range is: ； Where y(3000) is the absorbance value at wave number 3000; y(2200) is the absorbance value at wave number 2200; x(3000) is the abscissa at wave number 3000; x(2200) is the abscissa at wave number 2200; yi is the absorbance value after processing from wave number 2200 to 3000; D2=[y(2200),yi,……,y(3000)]; D = {D1, D2, D3}; E={D1,D2,D3}; D1 is 1200-2200cm in data D -1 Datasets; D2 is 2200-3000cm in data D -1 Datasets; D3 is 3000-4000cm in data D -1 Dataset.

6. The noise processing method for surface enhanced in-situ electrochemical infrared data according to claim 1, characterized in that: In S5, the deduction calculation formula is: X = F + κ * E; The spectral data X is [X1, X2, X3…Xm]; κ is the reduction factor.

7. The noise processing method for surface enhanced in-situ electrochemical infrared data according to claim 6, characterized in that: The deduction coefficient κ is in the range of [-100100], with an interval of 0.

01.

8. The noise processing method for surface enhanced in-situ electrochemical infrared data according to claim 6, characterized in that: The completion of the subtraction calculation is determined by the size of the standard deviation σ of the spectral data X. The calculation method of the standard deviation σ is as follows: ； Y=[X(2)X(1),X(3)X(2),…,X(m)X(m1)]; σ represents the standard deviation, yi represents each element in the data set Y, μ represents the mean value of the data set, m-1 represents the number of elements in the dataset.

9. The noise processing method for surface enhanced in-situ electrochemical infrared data according to any one of claims 6 or 7, characterized in that: Select 3500-4000cm -1 The standard deviation σ within the band range is used as the standard to judge whether the total intensity value of the spectrum after subtraction reaches the minimum value. When the total value reaches the minimum, the subtraction is considered to be completed.

Images