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

Abstract

This invention relates to the field of electrochemical experimental technology, specifically to a noise processing method for surface-enhanced in-situ electrochemical infrared data. By introducing 99.99% pure nitrogen gas into the infrared spectrometer to remove water vapor, and automatically and continuously acquiring spectral information at different voltages within the range of -0.8 to 0.8 V at 0.05 V intervals, until the carbon dioxide double peak intensity is between 2240 and 2400 cm⁻¹, the method continues until the carbon dioxide double peak intensity is between 2240 and 2400 cm⁻¹. ‑1 The peak size decreases until it almost disappears; the last two background data points A and B collected in the above spectral information are converted to obtain the infrared spectrum C of water vapor; the infrared spectrum C is normalized by maximum and minimum data to obtain the normalized water vapor peak data D; the water vapor peak data D is then divided into peaks at 2200-3000 cm⁻¹. ‑1 The carbon dioxide peak is subtracted from the range to obtain the water vapor infrared spectrum E. The absorbance F of the infrared spectrum data to be subtracted and the numerical matrix E of the water vapor spectral intensity are imported into MATLAB 2023B software. After the subtraction calculation is performed, the subtracted infrared spectrum data is output. This method is more efficient and accurate than the traditional manual subtraction.

Description

Technical Field

[0001] This invention relates to the field of electrochemical experimental technology, and more specifically to a noise processing method for surface-enhanced in-situ electrochemical infrared data. Background Technology

[0002] Surface-enhanced electrochemical in-situ infrared spectroscopy (SIR) holds significant scientific and practical value, enabling real-time monitoring of molecular reactions on electrode surfaces and providing direct evidence for studying molecular reaction mechanisms and kinetic processes. Through surface enhancement, this technology significantly improves the detection capability for low-concentration molecules and weak signals, effectively distinguishing minute changes in electrochemical reactions and providing molecular structural information, thus deepening our understanding of the surface chemistry and physical properties of materials. With the deepening of research in fields such as catalysis and energy storage, in-situ infrared spectroscopy is receiving increasing attention due to its potential in monitoring complex reactions, studying reaction mechanisms, and discovering intermediate products.

[0003] However, in practical applications, the large amount of infrared spectral data generated is often affected by water vapor, resulting in weak signals that are difficult to interpret. Generally, expensive methods such as vacuum infrared spectroscopy and long-term nitrogen purging 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 the impact of these interferences.

[0004] Traditional water vapor deduction methods primarily rely on manually adjusting the difference spectrum, calculating the difference spectrum using instrumental software (such as Thermo Fisher Scientific's OMNIC). Researchers need to perform multiple trials (typically 10-100 times) to obtain a suitable deduction effect. This method is not only time-consuming and labor-intensive but also prone to errors due to human judgment, significantly reducing the reliability of the analytical results.

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

[0006] To address the issue of noise interference from water vapor in infrared spectral data, traditional subtraction methods rely heavily on manual adjustments, resulting in low efficiency, large errors, and weak signals. Therefore, this invention provides a noise processing method for surface-enhanced in-situ electrochemical infrared data. The specific steps of this method are as follows:

[0007] S1. Introduce 99.99% pure nitrogen gas into the infrared spectrometer to remove water vapor and carbon dioxide, and automatically and continuously acquire spectral information at different voltages within the range of -0.8V to 0.8V at 0.05V intervals until the carbon dioxide double peak intensity is between 2240-2400cm². -1 Shrinking to the point of almost disappearing;

[0008] S2. The background data collected in the last two spectral data sessions above. and The conversion yields the infrared spectrum of water vapor. ;

[0009] S3. Regarding the above infrared spectrum Perform max-min normalization on the data to obtain normalized water vapor peak data. ;

[0010] S4. The above water vapor peak data At 2200-3000cm -1 Subtracting the carbon dioxide peak from the range yields the water vapor infrared spectrum. ;

[0011] S5. The absorbance of the infrared spectral data to be deducted. and the numerical matrix of water vapor spectral intensity Import In the software, according to After performing the deduction calculation, output the infrared spectral data after deduction. ;

[0012] Among them, infrared spectral data for , This is a deduction factor.

[0013] Preferably, in step S1, the frequency of automatically and continuously collecting spectral information under different voltages is once every 120 seconds.

[0014] Preferably, in step S2, the background data , The conversion relationship with the infrared spectrum is as follows: .

[0015] Preferably, in step S3, the infrared spectrum The maximum and minimum normalization range for the data is 0-0.005, and the specific normalization formula is as follows:

[0016] ;

[0017] in, It is the raw data;

[0018] It is the minimum value in the original data;

[0019] It is the maximum value in the original data;

[0020] It is normalized water vapor peak data.

[0021] Preferably, in step S4, the water vapor peak data At 2200-3000cm -1 The formula for subtracting the carbon dioxide peak within the range is:

[0022] ;

[0023] in, This is the absorbance value at a wave value of 3000;

[0024] This is the absorbance value at a wave value of 2200;

[0025] The horizontal axis wave value is 3000;

[0026] The horizontal axis wave value is 2200;

[0027] The absorbance values ​​after processing at wavenumbers of 2200 to 3000;

[0028] The value is the wave value;

[0029] ;

[0030] ;

[0031] ;

[0032] It is data 1200-2200cm -1 Dataset;

[0033] It is data 2200-3000cm -1 Dataset;

[0034] It is data 3000-4000cm -1 Dataset.

[0035] Preferably, the deduction coefficient Within the range of [-100 100], the interval is 0.01.

[0036] Preferably, through spectral data Standard deviation The size of the standard deviation is used to determine whether the deduction calculation is complete. The calculation method is as follows:

[0037] ;

[0038] ;

[0039] Represents standard deviation;

[0040] Representative dataset Each element in;

[0041] Represents the average value of the dataset;

[0042] This represents the number of elements in the dataset.

[0043] Preferably, a diameter of 3500-4000cm is selected. -1 Standard deviation within the band range As a standard for judging whether the total spectral intensity value after deduction has reached the minimum value, the deduction is considered to be complete when the total value reaches the minimum.

[0044] Compared with the prior art, the present invention has the following advantages:

[0045] This invention differs from traditional manual adjustment methods by utilizing the infrared signal of water vapor, which is mainly located in the 3500-4000 cm⁻¹ range of infrared spectra. -1 Because it operates within a specific spectral band, it avoids mistakenly subtracting data from other organic compounds during the subtraction process. Through iterative calculations, it obtains the minimum standard deviation corresponding to the smoothest waveform within that band, thereby selecting the infrared spectrum with the lowest total spectral intensity value. This method avoids the tedious process of manually adjusting coefficients, reducing subtraction time to a few seconds and significantly improving data processing efficiency. It is particularly suitable for processing large batches of infrared spectral data, and the algorithm can automatically determine whether subtraction is complete, ensuring the accuracy of spectral subtraction. Attached Figure Description

[0046] Figure 1 This is a schematic diagram showing the gradual decrease in the intensity of the bimodal peaks of water vapor and carbon dioxide during the data collection phase of this invention.

[0047] Figure 2 The infrared spectrum of water vapor in the experimental example of this invention;

[0048] Figure 3 This is the water vapor infrared spectrum after background drift was removed in the experimental examples of this invention;

[0049] Figure 4 The infrared spectrum of the test examples of this invention contains a large number of interfering peaks;

[0050] Figure 5This is the infrared spectrum after artificially subtracting water vapor in the experimental examples of this invention;

[0051] Figure 6 The infrared spectrum obtained by applying the subtraction algorithm of this invention in the experimental example of this invention. Detailed Implementation

[0052] To enhance understanding of the present invention, the present invention will be further described in detail below with reference to embodiments. These embodiments are only used to explain the present invention and do not constitute a limitation on the scope of protection of the present invention.

[0053] First, data acquisition was performed. The in-situ electrochemical device was installed, the optical path was adjusted, and 99.99% pure nitrogen gas 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 Scientific IS50 infrared spectrometer was used to automatically and continuously acquire spectra, collecting background data every 120 seconds until the range of 2240-2400 cm⁻¹ was reached. -1 Within the range, the intensity of the carbon dioxide bimodal peaks decreases to almost disappear, see Figure 2 As shown.

[0054] Because infrared background data is acquired using a single beam, while ordinary infrared spectra are presented in absorbance terms, and absorbance is obtained by logarithmically transforming the single-beam data, the infrared spectral data needs to be divided by the background (A / B) to obtain the spectrum. This is followed by a logarithmic transformation to obtain the absorbance map. The specific transformation formula is as follows: Furthermore, in order to remove the influence and interference of the carbon dioxide bimodal intensity on the spectrum, , These are the two last background data collected during the data collection phase described above.

[0055] Then, the above data The maximum and minimum normalized values ​​are between 0 and 0.005. The formula is as follows:

[0056] ;

[0057] in, It is the raw data;

[0058] It is the minimum value in the original data;

[0059] It is the maximum value in the original data;

[0060] It is normalized water vapor peak data.

[0061] Then, the water vapor peak data At 2200-3000cm -1By subtracting the carbon dioxide peak and its interference within the range, the water vapor infrared spectrum is obtained. The specific formula is as follows:

[0062] ;

[0063] in, This is the absorbance value at a wave value of 3000;

[0064] This is the absorbance value at a wave value of 2200;

[0065] The horizontal axis wave value is 3000;

[0066] The horizontal axis wave value is 2200;

[0067] The absorbance values ​​after processing at wavenumbers of 2200 to 3000;

[0068] The value is the wave value;

[0069] ;

[0070] ;

[0071] ;

[0072] It is data 1200-2200cm -1 Dataset;

[0073] It is data 2200-3000cm -1 Dataset;

[0074] It is data 3000-4000cm -1 Dataset.

[0075] Finally, the absorbance of the infrared spectral data to be deducted and the numerical matrix of water vapor spectral intensity Import The deduction is calculated according to the following formula:

[0076] ;

[0077] Infrared spectral data for , This is a deduction factor.

[0078] Since the original data signal intensity is around 0.01-0.1, while the water vapor signal intensity is around 0.005, it can be seen that the original data signal intensity is 2-20 times that of the water vapor signal. Furthermore, since the signal can be positive or negative, multiplying the water vapor spectral intensity value by [-100 to 100] is a suitable range, i.e., a subtraction factor. The value range is [-100 100]. To avoid excessive data processing affecting deduction efficiency, or insufficient data processing failing to achieve optimal water vapor deduction, the deduction coefficient is set to... The selection interval is set to 0.01.

[0079] Thus, when the absorbance of infrared spectral data After the subtraction calculation, the spectral dataset will be obtained. However, finding the optimal subtracted data from the extensively subtracted spectral data presents a new challenge. Therefore, in addition to the subtraction calculations, the standard deviation needs to be introduced. The calculation.

[0080] Specific standard deviation The calculation method is as follows:

[0081] ;

[0082] ;

[0083] Represents standard deviation;

[0084] Representative dataset Each element in;

[0085] Represents the average value of the dataset;

[0086] This represents the number of elements in the dataset.

[0087] Since the infrared spectral signal is mainly concentrated in the 1200-3500 cm⁻¹ range... -1 The wavelength range is 3500-4000cm. -1 Within this range, apart from the free hydroxyl peak of water vapor, most organic compounds do not produce signals in this region. The infrared signal of water vapor is mainly in the 1300-2000 cm⁻¹ range. -1 and 3500-4000cm -1 Wavelength band. Therefore, in the 3500-4000cm range. -1Within this band, if the spectral curve is smooth enough, it means that the water vapor peak is subtracted more thoroughly. In addition, since there are no other organic spectral signals in this band, there is no need to worry about missubtraction. This ensures that water vapor signal interference is subtracted while retaining the infrared signal of the electrochemical reaction, thus guaranteeing the accuracy of spectral signal subtraction.

[0088] Therefore, when we are at 3500-4000cm -1 Within this band, the minimum standard deviation is automatically selected by computer. The corresponding spectral data Then the spectral data It is the spectral data with the smallest total intensity value after deduction, which is the spectral data we need after deduction.

[0089] Experimental example:

[0090] In the experiment of platinum catalyst plating on gold-plated silicon crystal surface, nitrogen gas with a purity of 99.99% and a flow rate of 500 ml / min was first introduced into the instrument to purge water vapor and carbon dioxide. Single-beam spectra were collected every two minutes until the carbon dioxide peak basically disappeared.

[0091] The last two background data and Use the following formula to process: The infrared spectrum of water vapor was obtained, such as Figure 3 As shown. Then, normalization and carbon dioxide peak subtraction were performed to obtain the water vapor infrared spectrum, as shown. Figure 4 As shown.

[0092] Methanol was added to the system, and a surface-enhanced cyclic voltammetry electrochemical experiment was conducted with a voltage range of -0.8V to 0.8V and a sampling interval of 0.05V.

[0093] In the experiment, 32 sets of data containing numerous interfering peaks (such as water vapor) were obtained. Figure 5 The infrared spectral data shown. These abundant water vapor peaks severely interfere with spectral analysis, and traditional manual analysis methods were used to subtract each of the above 32 spectra, which took approximately 2 hours.

[0094] By using the noise processing method provided by this invention, researchers imported experimental data and Figure 2 The data was set, initial parameters were defined, and the number of iterations was set. After multiple iterations, the algorithm successfully subtracted interference peaks and output a clear infrared spectrum with minimal interference in approximately one second. Figure 6 .

[0095] The experimental results above demonstrate that the noise processing method of this invention significantly reduces the time required for manual adjustments and avoids errors caused by subjective judgment. The algorithm's automated processing capability is particularly outstanding when handling large-scale data, significantly improving 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 gas into the infrared spectrometer to remove water vapor and carbon dioxide, and automatically and continuously acquire spectral information at different voltages within the range of -0.8V to 0.8V at 0.05V intervals until the carbon dioxide double peak intensity is between 2240-2400cm². -1 Shrinking to the point of almost disappearing; S2. The background data collected in the last two spectral data sessions above. and The conversion yields the infrared spectrum of water vapor. ; S3. Regarding the above infrared spectrum Perform max-min normalization on the data to obtain normalized water vapor peak data. ; S4. The above water vapor peak data At 2200-3000cm -1 Subtracting the carbon dioxide peak from the range yields the water vapor infrared spectrum. ; S5. The absorbance of the infrared spectral data to be deducted. and the numerical matrix of water vapor spectral intensity Import In the software, according to After performing the deduction calculation, output the infrared spectral data after deduction. ; Among them, infrared spectral data for , This is a deduction factor; Through spectral data Standard deviation The size of the standard deviation is used to determine whether the deduction calculation is complete. The calculation method is as follows: ； ； Represents standard deviation; Representative dataset Each element in; Represents the average value of the dataset; Represents the number of elements in the dataset; Select 3500-4000cm -1 Standard deviation within the band range As a standard for judging whether the total spectral intensity value after deduction has reached the minimum value, the deduction is considered to be complete when the total value reaches the minimum.

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, background data , The conversion relationship with the infrared spectrum is as follows: .

4. The noise processing method for surface-enhanced in-situ electrochemical infrared data according to claim 1, characterized in that, In S3, the infrared spectrum The maximum and minimum normalization range for the data is 0-0.005, and the specific normalization formula is as follows: ； in, It is the raw data; It is the minimum value in the original data; It is the maximum value in the original data; It is 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 At 2200-3000cm -1 The formula for subtracting the carbon dioxide peak within the range is: ； in, This is the absorbance value at a wave value of 3000; This is the absorbance value at a wave value of 2200; The horizontal axis wave value is 3000; The horizontal axis wave value is 2200; The absorbance values ​​after processing at wavenumbers of 2200 to 3000; The value is the wave value; ； ； ； It is data 1200-2200cm -1 Dataset; It is data 2200-3000cm -1 Dataset; It is data 3000-4000cm -1 Dataset.

6. The noise processing method for surface-enhanced in-situ electrochemical infrared data according to claim 1, characterized in that, The deduction factor Within the range of [-100 100], the interval is 0.01.

Images