Raman spectrum quantitative analysis method taking water Raman spectrum as interior label

Abstract

The invention discloses a Raman spectrum quantitative analysis method taking a water Raman spectrum as an internal standard, and the Raman spectrum quantitative analysis method taking the water Raman spectrum as the internal standard is different from a conventional solution system quantitative analysis method in which a characteristic peak is selected as an internal standard peak. The spectrogram of the whole water is selected as an interior label, a normalization coefficient is calculated in a mode of perfectly deducting the spectrogram of the water, and the target of accurate quantitative analysis is achieved. The two-order derivative standard deviation of the differential spectrum base line is used as a criterion, serious interference of noise in the Raman spectrum on the two-order derivative of the differential spectrum is avoided, and meanwhile, interference of fluorescence on quantitative analysis results is effectively avoided in a mode of obtaining the pure solvent spectrum by deducting the fitting base line.

Description

Technical Field

[0001] The present application relates to a Raman spectroscopy quantitative analysis method using water Raman spectrum as an internal standard, belonging to the technical field of Raman spectroscopy material detection. Background Art

[0002] Raman spectroscopy can analyze the molecular vibration and rotational energy level information inside a substance, is highly sensitive to molecular structure, and has the characteristics of non-contact measurement, fast detection speed, and no need for special sample preparation. Therefore, Raman spectroscopy has been widely used in processes such as structural detection and component analysis. In addition, Raman spectroscopy can also be used for quantitative analysis. However, since the intensity of the Raman signal is affected by laser power, instrument response efficiency, measurement error, etc., it is necessary to select an external standard or an internal standard to normalize the Raman spectrum intensity during the quantitative analysis process to achieve quantitative analysis.

[0003] The external standard method often refers to selecting a characteristic Raman peak outside the sample as the standard, such as selecting the Raman peak of nitrogen in the air as the external standard. This method cannot eliminate the impact of measurement errors and requires a high degree of sample position overlap in each measurement. It is rarely used now. The internal standard method usually selects a Raman characteristic peak of a component in the sample itself as the standard, such as water at 1640cm in aqueous solution. -1 The vibration peak of the sample is used as the internal standard. Although this method can achieve quantitative analysis, it is very dependent on the stability of the internal standard peak. In the complex environment of actual application, such as reaction solution or culture medium, there is often a lack of sufficiently stable internal standard peaks, making it difficult to perform accurate quantitative analysis. Summary of the invention

[0004] In view of the problems faced in Raman spectroscopy quantitative analysis, the present invention applies for a Raman spectroscopy quantitative analysis method using water Raman spectrum as internal standard, avoiding the interference of one Raman peak as internal standard on quantitative analysis, and at the same time reducing the influence of noise and fluorescence background in Raman spectrum on analysis results. To achieve the above objectives, the present invention adopts the following technical solutions:

[0005] According to one aspect of the present application, a Raman spectroscopy quantitative analysis method using water Raman spectrum as an internal standard is provided, and the Raman spectroscopy quantitative analysis method comprises the following steps:

[0006] Step 1: Collect Raman spectra of water and sample solution, denoted as S 溶液 and S 溶剂 ;

[0007] Step 2: According to S 溶液 , S 溶剂 and calibration coefficient C, and obtain the difference spectrum S;

[0008] Step 3, fitting the difference spectrum S to obtain the corresponding baseline B, and calculating the standard deviation V of the second-order derivative of the baseline B;

[0009] Step 4: Determine whether V meets the convergence condition: If not, increase the calibration coefficient C and return to step 2; If it meets the convergence condition, output the calibration coefficient C0 at this time as the ideal calibration coefficient and the corresponding baseline B0 to obtain the normalized pure solute spectrum S 溶质 ;

[0010] Step 5: Spectrum of pure solute S 溶质 Perform peak fitting to obtain the normalized peak intensity I0 of the sample to be tested;

[0011] Step 6, according to the above steps 1 to 5, obtain a calibration curve of solute concentration and solute peak intensity;

[0012] Step 7: Calculate the concentration of the sample to be tested using the peak intensity I0 of the sample to be tested and the calibration curve.

[0013] Optionally, in step 2, the initial value of the calibration coefficient C is set to 0.001.

[0014] Optionally, in step 2, the calculation formula of the difference spectrum S is:

[0015] S=S 溶液 -S 溶剂 *C

[0016] S 溶液 is the Raman spectrum of the sample solution to be tested, S 溶剂 is the Raman spectrum of water, and C is the calibration coefficient.

[0017] Optionally, in step 3, the fitting adopts an asymmetric weighted penalty least squares method to fit the baseline, specifically:

[0018] (W+λD T D) B=WS

[0019] So solving the equation gives the baseline B:

[0020] B=(W+λD T D) -1 WS

[0021] Among them, W is the diagonal matrix composed of weight factors, λ is the smoothing factor, D is the difference matrix, S is the difference spectrum, and D T is the transposed matrix of the difference matrix.

[0022] Optionally, in step 4, the convergence condition is:

[0023] V i+1 <V i <Vi-1 ;

[0024] V i is the standard deviation V of the second-order derivative obtained in the i-th cycle;

[0025] If the convergence condition is met, the output V i The corresponding calibration coefficient is taken as the ideal calibration coefficient C0. If the convergence condition is not met, the calibration coefficient C is increased by 0.001 and the process returns to step 2.

[0026] Optionally, in step 4, the normalized pure solute mass spectrum S 溶质 The calculation formula is:

[0027]

[0028] Among them, S 溶液 , S 溶剂 , calibration coefficient C0, baseline B0, S 溶质 The same as S in claim 1 溶液 , S 溶剂 , calibration coefficient C0, baseline B0, S 溶质 Consistent.

[0029] Optionally, in step 5, the peak separation fitting method includes at least one of Gaussian fitting, Lorentz fitting, and Voigt function fitting.

[0030] Optionally, in step 1, the sample solution to be tested is a pre-prepared sample solution to be tested with a known concentration, and at least five sample solutions to be tested are prepared.

[0031] Optionally, in step 6, a series of sample solutions with known concentrations are first prepared (no less than five), and then the peak intensities I0 corresponding to different concentrations are obtained according to steps 1-5, and a calibration curve is obtained by linear fitting.

[0032] The beneficial effects of this application include:

[0033] 1) When the present application performs quantitative analysis on the solution system, the spectrum of the entire water is selected as the internal standard, and the normalization coefficient is calculated by perfectly subtracting the spectrum of water to achieve the goal of accurate quantitative analysis. The standard deviation of the second-order derivative of the difference spectrum baseline is used as the criterion to avoid the serious interference of noise in the Raman spectrum on the second-order derivative of the difference spectrum. At the same time, by subtracting the fitting baseline to obtain the pure solvent spectrum, the interference of fluorescence on the quantitative analysis results is effectively avoided.

[0034] 2) This application uses an algorithm to process actual data multiple times, and according to the effect of baseline correction, repeatedly corrects and determines the step size of the calibration coefficient, which can achieve the best quantitative analysis effect. If the step size of the calibration coefficient is reduced, such as changing "calibration coefficient increase by 0.001" to "calibration coefficient increase by 0.0001", the algorithm calculation time will be increased by more than 10 times, but there is no obvious difference in accuracy. If "calibration coefficient increase by 0.001" is changed to "calibration coefficient increase by 0.01", the solvent spectrum of water cannot be well deducted, resulting in a large error in the quantitative analysis results. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a flow chart of the method for quantitative analysis of Raman spectroscopy using water Raman spectrum as internal standard in this application.

[0036] Figure 2 The processing results in Example 1 of the present application are as follows: a is the original Raman spectra of 5 mM NaNO3 solution and water; b is the curve of the second-order derivative standard deviation V of the difference spectrum and the calibration coefficient C; c is the processed difference spectrum and the corresponding baseline B0.

[0037] Figure 3 NO3 in the NaNO3 standard solution in Example 1 of the present application - The calibration curve of

[0038] Figure 4 These are the Raman spectra of glucose at different concentrations after deducting the water Raman spectrum in Example 2 of the present application.

[0039] Figure 5 This is the calibration curve of glucose in Example 2 of the present application. DETAILED DESCRIPTION

[0040] The present application is described in detail below with reference to embodiments, but the present application is not limited to these embodiments.

[0041] Unless otherwise specified, the raw materials in the examples of this application were purchased through commercial channels.

[0042] According to two embodiments of the present invention, a method flow chart of a Raman spectroscopy quantitative analysis method for an aqueous solution system using water Raman spectrum as an internal standard is provided, such as Figure 1 As shown, the specific steps include:

[0043] Step 1: Collect the Raman spectra of the water solvent and the sample solution to be tested, denoted as S 溶液 and S 溶剂 ;

[0044] Step 2: According to S 溶液 and S 溶剂 and calibration coefficient C, and obtain the difference spectrum S;

[0045] The initial value of the calibration coefficient C is set to 0.001, and the difference spectrum S is obtained by direct subtraction, the formula is:

[0046] S=S 溶液 -S 溶剂 *C

[0047] Step 3, fitting the difference spectrum S to obtain the corresponding baseline B, and calculating the standard deviation V of the second-order derivative of the baseline B;

[0048] The baseline is fitted using an asymmetric weighted penalty least squares method, specifically:

[0049] (W+λD T D) B=WS

[0050] Among them, W is the diagonal matrix composed of weight factors, λ is the smoothing factor, D is the difference matrix, and D T is the transposed matrix of the difference matrix. Therefore, we can solve the following equation to get the baseline B:

[0051] B=(W+λD T D) -1 WS

[0052] Step 4: Determine whether V meets the convergence condition: If it does not meet the convergence condition, the calibration coefficient is increased by 0.001 and returns to step 2; If it meets the convergence condition, the calibration coefficient C0 at this time is output as the ideal calibration coefficient and the corresponding baseline B0 to obtain the normalized pure solute spectrum S 溶质 ;

[0053] The convergence condition is: V i+1 <V i <V i-1 . V i is the standard deviation V of the second derivative of the difference spectrum obtained in the i-th cycle;

[0054] Pure solute spectrum S 溶质 It is calculated by the following formula:

[0055]

[0056] Step 5: Spectrum of pure solute S 溶质 Perform peak fitting to obtain the normalized peak intensity I0 of the sample to be tested;

[0057] The peak fitting methods include but are not limited to Gaussian fitting, Lorentz fitting, Voigt function fitting, etc.

[0058] Step 6, obtaining a calibration curve of solute concentration and solute peak intensity according to the above steps 1 to 5;

[0059] First, prepare a series of sample solutions with known concentrations (no less than five), then obtain the peak intensities I0 corresponding to different concentrations according to steps 1-5, and obtain a calibration curve using linear fitting;

[0060] Step 7: Calculate the concentration of the sample to be tested using the peak intensity I0 of the sample to be tested and the calibration curve to complete the quantitative analysis.

[0061] The Raman spectroscopy quantitative analysis method provided in the present application, which uses the water Raman spectrum as the internal standard, is different from the traditional solution system quantitative analysis in which a characteristic peak is selected as the internal standard peak. The present application uses the solvent spectrum of water as the internal standard, and finds the normalization coefficient by subtracting the solvent spectrum from the solution spectrum, thereby accurately and quantitatively analyzing the concentration of the species to be measured.

[0062] The present invention provides two embodiments:

[0063] Example 1

[0064] 0.1 mM, 1 mM, 2 mM, 3 mM, 4 mM, and 5 mM sodium nitrate aqueous solutions were prepared respectively, and then Raman spectra were collected. Figure 2 a is the Raman spectrum of pure water and 5mM NaNO3 solution. It can be found that the absolute intensity difference between the two spectra is very large. This is due to factors such as measurement error. Therefore, the Raman spectrum needs to be normalized. According to the method proposed in this application, the difference spectrum under different calibration coefficients and the second-order derivative standard deviation V of the difference spectrum are calculated. For the convenience of display, Figure 2 b directly plots the relationship between the calibration coefficient C and V between 0.425 and 0.675, where when C = 0.562, the convergence condition V is met. i+1 <V i <V i-1 . Figure 2 c gives the difference spectrum and fitted baseline B0 at this time. When V is the smallest, that is, the calibration coefficient is 0.562, the difference spectrum no longer contains solvent spectrum information. Then the 1045cm -1 The normalized nitrate peak intensity is obtained by processing the Raman spectra of sodium nitrate aqueous solutions with different concentrations according to the above method, and the normalized peak intensity of nitrate at different concentrations is obtained, and the peak intensity and concentration curve is plotted, as shown in FIG. Figure 3 As shown, by linear fitting, the peak intensity and concentration equations were obtained, and the residual sum of squares was greater than 0.999, which showed a good linear relationship.

[0065] Example 2

[0066] 50, 150, 250, 350, and 450 mg / ml glucose aqueous solutions were prepared respectively, and then Raman spectra were collected. According to the processing flow of Example 1, the glucose aqueous solutions of different concentrations were processed to obtain glucose spectra after deducting water and normalization, as shown in FIG. Figure 4 . Select 1122cm -1 The Raman peak at is taken as the characteristic peak of glucose, and the equation of glucose peak intensity and concentration is obtained, see Figure 5 , the residual sum of squares is also greater than 0.999. This indicates that the method proposed in this application can accurately extract the Raman spectrum of the solute in the aqueous solution, and normalize the peak intensity, so as to perform quantitative analysis.

[0067] The above are only a few embodiments of the present application and do not constitute any form of limitation to the present application. Although the present application is disclosed as above with preferred embodiments, it is not intended to limit the present application. Any technician familiar with the profession, without departing from the scope of the technical solution of the present application, using the technical content disclosed above to make slight changes or modifications are equivalent to equivalent implementation cases and fall within the scope of the technical solution.

Claims

1. A Raman spectroscopy quantitative analysis method using water Raman spectrum as internal standard, characterized in that: The Raman spectroscopy quantitative analysis method comprises the following steps: Step 1: Collect Raman spectra of water and sample solution, denoted as S 溶液 and S 溶剂 ; Step 2: According to S 溶液 、S 溶剂 and calibration coefficient C, and obtain the difference spectrum S; Step 3: Fit the difference spectrum S to obtain the corresponding baseline B, and calculate the standard deviation V of the second-order derivative of the baseline B; Step 4: Determine whether V meets the convergence condition: If not, increase the calibration coefficient C and return to step 2; If it meets the convergence condition, output the calibration coefficient C0 at this time as the ideal calibration coefficient and the corresponding baseline B0 to obtain the normalized pure solute spectrum S 溶质 ; Step 5: Spectrum of pure solute S 溶质 Perform peak fitting to obtain the normalized peak intensity I0 of the sample to be tested; Step 6: According to the above steps 1 to 5, a calibration curve of solute concentration and solute peak intensity is obtained; Step 7: Calculate the concentration of the sample to be tested using the peak intensity I0 of the sample to be tested and the calibration curve.

2. The Raman spectroscopy quantitative analysis method according to claim 1, characterized in that: In step 2, the initial value of the calibration coefficient C is set to 0.

001.

3. The Raman spectroscopy quantitative analysis method according to claim 2, characterized in that: In step 2, the calculation formula of the difference spectrum S is: S=S 溶液 -S 溶剂 *C S 溶液 is the Raman spectrum of the sample solution to be tested, S 溶剂 is the Raman spectrum of water, and C is the calibration coefficient.

4. The Raman spectroscopy quantitative analysis method according to claim 1, characterized in that: In step 3, the fitting adopts an asymmetric weighted penalty least squares method to fit the baseline, specifically: (W+λD T D)B=WS Therefore, solving the equation gives the baseline B: B=(W+λD T D) -1 WS Among them, W is the diagonal matrix composed of weight factors, λ is the smoothing factor, D is the difference matrix, S is the difference spectrum, D T is the transposed matrix of the difference matrix.

5. The Raman spectroscopy quantitative analysis method according to claim 1, characterized in that: In step 4, the convergence condition is: In i+1 <V i <V i-1 ; V i is the standard deviation V of the second-order derivative obtained in the i-th cycle; If the convergence condition is met, the output V i The corresponding calibration coefficient is taken as the ideal calibration coefficient C0. If the convergence condition is not met, the calibration coefficient C is increased by 0.001 and the process returns to step 2.

6. The Raman spectroscopy quantitative analysis method according to claim 1, characterized in that: In step 4, the normalized pure solute mass spectrum S 溶质 The calculation formula is: Among them, S 溶液 、S 溶剂 , calibration coefficient C0, baseline B0, S 溶质 and S in claim 1 溶液 、S 溶剂 , calibration coefficient C0, baseline B0, S 溶质 consistent.

7. The Raman spectroscopy quantitative analysis method according to claim 1, characterized in that: In step 5, the peak fitting method includes at least one of Gaussian fitting, Lorentz fitting, and Voigt function fitting.

8. The Raman spectroscopy quantitative analysis method according to claim 1, characterized in that: In step 1, the sample solution to be tested is a sample solution to be tested with a known concentration prepared in advance, and at least five sample solutions to be tested are prepared.

Images