A new method to remove the interference of water vapor peak in infrared spectrum

Abstract

The present invention discloses a method for removing the interference of water vapor peaks in infrared spectra. Specifically, based on the concept of big data and the pigeonhole theory, a spectral library is established, and two single-beam background spectra that are closest to the peak position offset of the single-beam sample spectrum are selected from the spectral library. The three spectra are used to construct a two-dimensional asynchronous spectrum of the first-order derivative of the negative logarithm, and the spectrum of the pure substance in the two-dimensional asynchronous spectrum is extracted by intercepting the system missing peak of the two-dimensional asynchronous spectrum. This method for removing the interference of water vapor peaks in infrared spectra not only realizes the measurement and effective correction of the spectral offset between a given single-beam sample spectrum and a single-beam background spectrum, but also completely removes the interference of water in the gas phase through the two-dimensional asynchronous spectrum, and can successfully obtain a high-quality spectrum without the interference of water, as well as a reliable second-order derivative spectrum. The method described in the present invention provides a new idea for the study of fine spectral features contained in spectral regions that are severely interfered with and masked by water vapor.

Description

Technical Field

[0001] The invention belongs to the field of infrared spectrum analysis, and in particular relates to a method for removing interference of water vapor peaks in infrared spectrum. Background Art

[0002] In the past decades, FTIR spectroscopy has found significant applications in the fields of physical science, chemistry and materials research, medicine and pharmaceutical practice, and environmental protection. FTIR spectroscopy is a versatile tool for identifying unknown organic compounds and coordination compounds containing organic ligands. In addition, FTIR spectra can serve as sensitive probes to reflect structural changes in complex molecular systems in supramolecular assemblies, polymer materials, and biomolecular systems.

[0003] In many FTIR experiments, spectra are collected in an environment containing water. Absorption peaks caused by atmospheric water molecules in the infrared light path from the source to the detector appear in the final single-beam spectrum. Fluctuations of gaseous water during the measurement of the sample spectrum and the corresponding background spectrum can produce annoying interferences in the acquired absorption spectrum. The vibration-rotation peaks of gaseous water cover two broad spectral ranges in the mid-infrared spectrum, 4000-3000cm -1 and 2300~1300cm -1 Therefore, it will affect many important information vibration bands, such as OH stretching vibration, NH stretching vibration, C=O stretching vibration, C=C stretching vibration and CH 2 Bending vibration.

[0004] In the field of FTIR spectroscopy, some methods have been widely used to reduce the water vapor interference in the spectrum, such as removing the peak of gaseous water by spectral subtraction, but this method usually ends up producing positive and negative peaks at the same time, rather than completely eliminating the interference. One of the reasons is that temperature fluctuations cause the cavity of the helium-neon laser to expand and contract, resulting in peak shifts in the spectrum obtained by the FTIR spectrometer. Another example is smoothing the FTIR spectrum by the Savitzky-Golay algorithm or similar methods is an alternative method, but the interference of the vibration-rotation peak of gaseous water causes a series of peaks in the FTIR spectrum, whose characteristics are completely different from random noise, so the resulting spectrum is severely distorted.

[0005] In order to minimize distortion in spectral preprocessing, it is wise to treat the spikes as outliers. The prior art has proposed methods using spline approximation, polynomial filtering and excluding spectral data points contaminated by gaseous water. For example, Phillips and Harris introduced a missing point polynomial fitting method and a rigorous mathematical solution. Katsumoto and Ozaki provided a practical moving window algorithm to identify noise spikes caused by water vapor, then discarded the identified data points destroyed by water vapor, and then used interpolation techniques to reproduce the discarded data points. In the processed spectrum, the interference of water vapor was indeed significantly suppressed, but the above method has not been proven to be a reliable means for second-order derivative analysis of spectral bands with overlapping spectral peaks.

[0006] In view of the above reasons, it is urgent to develop an effective method to completely eliminate the interference of water from the original FTIR spectrum and the corresponding second-order derivative spectrum. Summary of the invention

[0007] In order to overcome the above problems, the inventors have conducted intensive research and developed a method for removing the interference of water vapor peaks in infrared spectra, which is called REACT-FILES method. Specifically based on the concept of big data and pigeonhole theory, a spectrum library is established, and two single-beam background spectra that are closest to the peak position offset of the single-beam sample spectrum are selected from the spectrum library. A two-dimensional asynchronous spectrum of the first-order derivative of the negative logarithm is constructed according to the single-beam sample spectrum and the single-beam background spectrum, and the spectrum of the pure substance in the two-dimensional asynchronous spectrum is extracted by intercepting the system peak of the two-dimensional asynchronous spectrum. This method for removing the interference of water vapor peaks in infrared spectra not only realizes the measurement and effective correction of the spectral offset between a given single-beam sample spectrum and a single-beam background spectrum, but also completely removes the interference of water in the gas phase through the two-dimensional asynchronous spectrum, and can successfully obtain a high-quality spectrum without the interference of water, as well as a reliable second-order derivative spectrum. The method described in the present invention provides a new idea for the study of fine spectral features contained in the spectral region that is severely interfered and masked by water vapor, thereby completing the present invention.

[0008] Specifically, the purpose of the present invention is to provide the following aspects:

[0009] In a first aspect, a method for removing interference from water vapor peaks in an infrared spectrum is provided, the method comprising:

[0010] Step 1, select two single-beam background spectra from the spectrum library;

[0011] Step 2, constructing a two-dimensional asynchronous correlation spectrum according to the single-beam sample spectrum and the single-beam background spectrum selected in step 1;

[0012] Step 3, processing the two-dimensional asynchronous correlation spectrum obtained in step 2 to obtain an infrared spectrum of the sample without water vapor interference.

[0013] Wherein, in step 1, the single-beam background spectrum is recorded and stored in a spectral library, which includes single-beam background spectra obtained from various seasons, various time periods, and different room temperature conditions, including single-beam background spectra of a helium-neon laser of an infrared spectrometer under different temperature states.

[0014] Wherein, in step 1, the single beam background spectrum selected from the spectrum library is determined according to the phase angle between the single beam background spectrum and the single beam sample.

[0015] Wherein, step 1 includes the following sub-steps:

[0016] Step 1-1, taking negative logarithms and derivatives of the single-beam sample spectrum and the single-beam background spectrum in the spectral library;

[0017] Step 1-2, using the derivative data obtained in step 1-1 to obtain the phase angle between the single beam sample spectrum and the single beam background spectrum;

[0018] Step 1-3, selecting two single-beam background spectra with the lowest phase angles according to the phase angles obtained in step 1-2.

[0019] Wherein, in step 1-1, the negative logarithm formula is as shown in Formula I:

[0020] L S (x) = -log 10 (B S (x)

[0021] L B (x) = -log 10 (B B (x)) Formula I;

[0022] Where: x is the wave number of infrared light;

[0023] B S (x) is the spectral intensity of the single beam sample spectrum Bs at the x wave number;

[0024] B B (x) is the single beam background spectrum B B Spectral intensity at wave number x;

[0025] L S (x) is the negative logarithm of the spectral intensity of the single beam sample spectrum Bs at the x wave number;

[0026] L B (x) is the single beam background spectrum B B Negative logarithm of the spectral intensity at wavenumber x.

[0027] Wherein, in step 1-1, the derivation formula is as shown in Formula II:

[0028]

[0029] Where: L' S (x) is the first derivative of the negative logarithm of the spectral intensity of the single-beam sample spectrum Bs at the peak position x;

[0030] L' B (x) is the single beam background spectrum B B The first derivative of the negative logarithm of the spectral intensity at the peak position x.

[0031] Wherein, in step 1-2, the phase angle calculation formula is shown in Formula III:

[0032]

[0033] Where: θ is the single beam sample spectrum L' S (x i ) and single beam background spectrum L' B (x i ) between the phase angles;

[0034] x i 4000cm -1 ~3750cm -1 Spectral data points between 3855cm -1 ~3850cm -1 Spectral data points between;

[0035] L' S (x i ) is the single beam sample spectrum Bs at x i The first derivative of the negative logarithm of the spectral intensity at the spectral data point;

[0036] L' B (x i ) is the single beam background spectrum B B In x i The first derivative of the negative logarithm of the spectral intensity at the spectral data point.

[0037] Wherein, the step 2 comprises the following steps:

[0038] Step 2-1, converting the first-order derivative of the negative logarithm of the single-beam sample spectrum and the two selected single-beam background spectra into a dynamic spectrum;

[0039] Step 2-2, constructing the dynamic spectrum obtained in step 2-1 into a two-dimensional asynchronous correlation spectrum.

[0040] Wherein, in step 3, the processing method is to obtain a pure sample spectrum without water vapor interference by taking a line cut at the system missing peak in the two-dimensional asynchronous correlation spectrum and integrating the line cut.

[0041] In a second aspect, there is provided an application of the method described in the first aspect for removing interference from water vapor peaks in infrared spectra.

[0042] The beneficial effects of the present invention include:

[0043] (1) The method for removing water vapor peak interference in infrared spectrum provided by the present invention realizes the measurement and effective correction of the spectral offset between a given single-beam sample spectrum and a single-beam background spectrum.

[0044] (2) The method for removing water vapor peak interference in infrared spectrum provided by the present invention constructs a two-dimensional asynchronous spectrum of the first-order derivative of the negative logarithm according to a given single-beam sample spectrum and a single-beam background spectrum, which effectively eliminates the orthogonality destruction caused by non-zero baseline drift.

[0045] (3) The method for removing water vapor peak interference in infrared spectra provided by the present invention completely removes the interference of water in the gas phase through two-dimensional asynchronous spectroscopy, and can successfully obtain high-quality spectra without interference from water, as well as reliable second-order derivative spectra.

[0046] (4) The method for removing water vapor peak interference in infrared spectra provided by the present invention corrects the lateral deviation of the spectrum caused by the temperature fluctuation of the helium-neon laser cavity, effectively eliminates the water vapor peak, and can obtain multiple sample spectra without moisture interference. The average of these intercepts can improve the signal-to-noise ratio level of the obtained spectrum.

[0047] (5) The method for removing the interference of water vapor peaks in infrared spectra provided by the present invention provides a new idea for studying the fine spectral features contained in the spectral regions that are seriously interfered with and masked by water vapor. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A comparison diagram of the gas phase ethanol sample of Example 1 and the background spectrum in the background spectrum library using phase angle matching is shown;

[0049] Figure 2 The two-dimensional asynchronous correlation spectrum of negative logarithmic processing of gas-phase ethanol in Example 1 is shown;

[0050] Figure 3 The two-dimensional asynchronous correlation spectrum of the negative logarithmic first-order derivative processing of gas-phase ethanol in Example 1 is shown;

[0051] Figure 4 The spectra of different treatment processes of gas-phase ethanol in Example 1 are shown;

[0052] Figure 5The spectra of different treatment methods of gas-phase ethanol in Example 2 are shown;

[0053] Figure 6 The spectra obtained by different PE treatment methods in Example 3 are shown;

[0054] Figure 7 The spectra obtained by different treatment methods of EVA in Example 4 are shown. DETAILED DESCRIPTION

[0055] The present invention will be further described in detail below through the accompanying drawings and embodiments. Through these descriptions, the characteristics and advantages of the present invention will become more clear and distinct.

[0056] The word "exemplary" is used exclusively herein to mean "serving as an example, embodiment, or illustration." Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments. Although various aspects of the embodiments are shown in the drawings, the drawings are not necessarily drawn to scale unless otherwise noted.

[0057] The first aspect of the present invention aims to provide a method for removing interference from water vapor peaks in infrared spectra, the method being referred to as the REACT-FILES method, comprising:

[0058] Step 1: Select two single-beam background spectra from the spectral library.

[0059] In step 1, the single-beam background spectrum is recorded and stored in a spectral library, which is established based on the concept of big data and the pigeon hole theory, including single-beam background spectra from various seasons and time periods, including single-beam background spectra of infrared spectrometer helium-neon lasers in different temperature states.

[0060] In step 1, the single-beam background spectrum selected from the spectrum library is closest to the single-beam sample spectrum peak shift, preferably 2, wherein the peak shift between the selected single-beam background spectrum and the single-beam sample spectrum can be determined by the phase angle between the single-beam background spectrum and the single-beam sample. The smaller the phase angle, the closer the peak shift between the selected single-beam background spectrum and the single-beam sample spectrum is. This is because the phase angle change of the absorbance spectrum has nothing to do with the peak intensity, but only with the peak shift, and the phase angle is very sensitive to the slight band shift of the absorption peak. Therefore, the phase angle can be used to measure the spectral peak shift, and then the single-beam background spectrum closest to the peak shift of the single-beam sample spectrum is obtained. The specific operation is as follows:

[0061] Step 1-1, taking the negative logarithm and derivative of the single-beam sample spectrum and the single-beam background spectrum in the spectral library.

[0062] In step 1-1, the negative logarithm formula is as shown in Formula I:

[0063] L S (x) = -log 10 (B S (x)

[0064] L B (x) = -log 10 (B B (x)) Formula I;

[0065] Where: x is the wave number of infrared light;

[0066] B S (x) is the spectral intensity of the single beam sample spectrum Bs at the x wave number;

[0067] B B (x) is the single beam background spectrum B B Spectral intensity at wave number x;

[0068] L S (x) is the negative logarithm of the spectral intensity of the single beam sample spectrum Bs at the x wave number;

[0069] L B (x) is the single beam background spectrum B B Negative logarithm of the spectral intensity at wavenumber x.

[0070] In step 1-1, the derivation formula is shown in Formula II:

[0071]

[0072] Where: L' S (x) is the first derivative of the negative logarithm of the spectral intensity of the single-beam sample spectrum Bs at the x wavenumber;

[0073] L' B (x) is the single beam background spectrum B B The first derivative of the negative logarithm of the spectral intensity at wavenumber x.

[0074] Step 1-2, using the derivative data obtained in step 1-1 to obtain the phase angle between the single-beam sample spectrum and the single-beam background spectrum.

[0075] In step 1-2, the phase angle calculation formula is shown in Formula III:

[0076]

[0077] Where: θ is the single beam sample spectrum L' S (x i ) and single beam background spectrum L' B (x i ) between the phase angles;

[0078] L' S (x i ) is the single beam sample spectrum Bs at x i The first derivative of the negative logarithm of the spectral intensity at the spectral data point;

[0079] L' B (x i ) is the single beam background spectrum B B In x i The first derivative of the negative logarithm of the spectral intensity at the spectral data point;

[0080] Among them, x i 4000cm -1 ~3750cm -1 Spectral data points between 3855cm -1 ~3850cm -1 spectral data points between .

[0081] According to the present invention, B S (x) and B B (x) contains water vapor. The difference in water vapor between the two comes from two aspects: 1) the difference in the transient concentration of gaseous water in the optical path. 2) the temperature difference of the optical cavity of the HeNe laser. Due to the Brownian motion of water molecules in the air, the transient concentration of water in the optical path of the FTIR spectrometer is always in a fluctuating state. Therefore, when collecting single-beam sample spectra and single-beam background spectra, the difference in gaseous water concentration is inevitable. S (x) and B B The difference in moisture content at (x) will produce a series of interfering peaks related to gaseous water in the final absorption spectrum of the sample.

[0082] Theoretically, the moisture interference caused by the different transient concentrations of gaseous water can be eliminated by spectral subtraction. However, the temperature difference of the optical cavity of the HeNe laser often makes the spectral subtraction ineffective. In the measurement of the single-beam sample spectrum and the background spectrum, the slight fluctuation of the temperature of the HeNe laser cannot be avoided. It turns out that if the temperature fluctuation is at the level of 0.1℃, the corresponding spectral shift is about 0.03cm -1 Therefore, even slight temperature fluctuations may cause systematic spectral shifts in the obtained single-beam spectrum. Although it is difficult to see such subtle changes in the observed spectrum, they are sufficient to invalidate all spectral subtraction-based methods. In the prior art, there is no report on methods for eliminating interference caused by temperature fluctuations on the optical cavity of HeNe lasers.

[0083] The inventors have found that the phase angle change of the absorbance spectrum has nothing to do with the peak intensity, but is only related to the peak position shift, and the phase angle is very sensitive to the slight band shift of the absorption peak. Therefore, the present invention uses the phase angle to measure the spectrum shift. Based on the value of the phase angle, a single-beam background spectrum with the lowest phase angle is selected. When the single-beam sample spectrum and the selected single-beam background spectrum generate an absorption spectrum, the interference caused by the system offset related to the optical cavity temperature fluctuation of the helium-neon laser of the FTIR spectrometer can be eliminated.

[0084] According to the present invention, if the phase angle between two spectra is calculated based on the entire spectral region, the spectral signal of the sample will produce a non-zero phase angle, making the phase angle invalid in reflecting the spectral shift. The inventors have found that the 4000cm -1 ~3750cm -1 The signal between 4000cm and 4000cm contains only the contribution of gaseous water. In order to avoid the judgment of the phase angle by the absorption peak of the single-beam sample spectrum, only -1 ~3750cm -1 The phase angle is calculated by using the spectral data between

[0085] According to the present invention, at 4000cm -1 ~3750cm -1 There are multiple peaks related to gaseous water in the spectral region between 3855cm and 4856cm. Each peak is related to a specific vibration-rotation energy level transition. Different air temperatures can cause different populations of water molecules at different rotational energy levels, resulting in differences in the relative intensities of different vibration-rotation peaks of gaseous water between the two spectra. The differences in the relative intensities of different vibration-rotation peaks may produce undesirable non-zero phase angles. An effective way to solve this problem is to use the spectral data within a single vibration-rotation peak of gaseous water to calculate the phase angle. -1 ~3850cm -1 The vibration-rotation peak of gaseous water is the strongest, therefore, in a further preferred embodiment, 3855 cm -1 ~3850cm -1 Calculate the phase angle between the spectral data.

[0086] Step 1-3, selecting two single-beam background spectra with the lowest phase angles according to the phase angles obtained in step 1-2.

[0087] Step 2: construct a two-dimensional asynchronous correlation spectrum based on the single-beam sample spectrum and the single-beam background spectrum selected in step 1.

[0088] In step 2, preferably, a two-dimensional asynchronous spectrum is established based on the two single-beam background spectra with the lowest phase angles selected in step 1 and the single-beam sample spectrum. At this time, the temperature difference of the optical cavity of the helium-neon laser and the slight changes on the moisture peak have been eliminated. The specific steps are as follows:

[0089] Step 2-1, converting the first-order derivative of the negative logarithm of the single-beam sample spectrum and the two selected single-beam background spectra into a dynamic spectrum.

[0090] In step 2-1, the spectrum average is calculated by formula IV, and the spectrum is converted into a dynamic spectrum by formula V.

[0091] According to the present invention, formula IV is shown below:

[0092]

[0093] Where: x is the wave number of infrared light;

[0094] L' S (x) is the first derivative of the negative logarithm of the spectral intensity of the single-beam sample spectrum Bs at the x wavenumber;

[0095] L' Bm (x) is the single beam background spectrum B with the lowest phase angle Bm The first derivative of the negative logarithm of the spectral intensity at wave number x;

[0096] L' Bn (x) is the single beam background spectrum B with the lowest phase angle Bn The first derivative of the negative logarithm of the spectral intensity at wave number x;

[0097] L' AV (x) represents the average value of the negative logarithmic first-order derivative of the spectral intensity at the x peak of the three single-beam spectra.

[0098] According to the present invention, formula V is shown below:

[0099]

[0100] Where: x is the wave number of infrared light;

[0101] It represents the dynamic spectral intensity of the negative logarithmic first derivative of the spectral intensity of the single beam sample spectrum Bs at the x wave number;

[0102] The single beam background spectrum B represents the lowest phase angle Bm Dynamic spectral intensity of the negative logarithmic first derivative of the spectral intensity at wave number x;

[0103] The single beam background spectrum B represents the lowest phase angle Bn Dynamic spectral intensity of the negative logarithmic first derivative of the spectral intensity at wavenumber x.

[0104] Step 2-2, constructing the dynamic spectrum obtained in step 2-1 into a two-dimensional asynchronous correlation spectrum.

[0105] In step 2-2, specifically, the dynamic spectrum obtained in step 2-1 is used to construct the matrix D according to formula VI 1 , and then generate a two-dimensional asynchronous correlation spectrum through formula VII.

[0106] According to the present invention, formula VI is as follows:

[0107]

[0108] Where: D 1 express Matrix of

[0109] x is the wave number of infrared light;

[0110] Indicates that the single beam sample spectrum Bs is at x 1 、x 2 ……x Q Dynamic spectral intensity of the negative logarithmic first derivative at wavenumber;

[0111] The single beam background spectrum B represents the lowest phase angle Bm Respectively in x 1 、x 2 ……x Q Dynamic spectral intensity of the negative logarithmic first derivative at wavenumber;

[0112] The single beam background spectrum B represents the lowest phase angle Bn Respectively in x 1 、x 2 ……x Q Dynamic spectral intensity of the negative logarithmic first derivative at wavenumber.

[0113] According to the present invention, formula VII is as follows:

[0114]

[0115] Where: L (x,y) represents the intensity at the midpoint (x,y) in the two-dimensional asynchronous correlation spectrum;

[0116] T represents the transpose of the matrix;

[0117] N represents the Hilbert-Noda matrix, which can be expressed as:

[0118]

[0119] According to the present invention, in step 2, a database is established based on big data and pigeonhole theory, and a two-dimensional asynchronous correlation spectrum is constructed by the first-order derivative of the negative logarithm of the single-beam spectrum with base 10. It has been proved that through phase angle matching, this method can effectively remove the interference of the single-beam sample spectrum and the single-beam background spectrum caused by the temperature fluctuation of the helium-neon laser cavity of the infrared spectrometer.

[0120] Step 3, processing the two-dimensional asynchronous correlation spectrum obtained in step 2 to obtain an infrared spectrum of the sample without water vapor interference.

[0121] In step 3, the processing method is to obtain a pure sample spectrum without water vapor interference by taking a line cut at the system missing peak in the two-dimensional asynchronous correlation spectrum and integrating the line cut.

[0122] The inventors have found that by taking a line segment from the peak-missing position of the two-dimensional system of the independent peak of water vapor and integrating the line segment, a pure sample spectrum without water vapor interference can be obtained.

[0123] According to the present invention, when Ψ L When (x, y) = 0, it means that the intensity at this point is zero, that is, there is no cross peak at (x, y) in the two-dimensional asynchronous spectrum, and this is a system missing peak.

[0124] According to the present invention, at 4000cm in the infrared spectrum, -1 ~3750cm -1 The band is only contributed by gaseous water, so 4000cm -1 ~3750cm -1 There will be a lot of missing peaks in the system, so the wavelength range of the intercept position is selected to be 4000cm -1 ~3750cm -1 After the selected wave number between the two, the spectrum will be normalized after the intercept integration, and the spectrum at different intercept positions will be obtained by normalization. According to the spectrum in the wavelength range of 4000cm -1 ~3800cm -1 The standard deviation between them is used to obtain the pure sample spectrum, wherein the spectrum corresponding to the minimum standard deviation is the pure sample spectrum. Preferably, the spectrum obtained by normalization processing is averaged to improve the signal-to-noise ratio, so as to obtain a higher quality pure sample spectrum without water vapor interference.

[0125] In step 3, at the systematic missing peak of the two-dimensional asynchronous correlation spectrum, along y=y 0 The pure sample spectrum without water vapor interference is obtained by integration, where y 0 Indicates that along the spectrum peak at 4000cm -1 ~3750cm -1 Any peak position whose absolute value of the first-order derivative is greater than 0.3.

[0126] According to the present invention, the integral formula is as shown in Formula VIII:

[0127] f S (x) = ∫Ψ L (x,y 0 )dx Formula VIII;

[0128] Where: L (x,y 0 ) represents the midpoint (x,y) in the two-dimensional asynchronous correlation spectrum. 0 ) at the point of strength;

[0129] y 0 is the intercept position point, x is the wave number of infrared light;

[0130] f S (x) represents the spectral factor at wave number x in the dimensional asynchronous correlation spectrum.

[0131] In the prior art, the second-order derivative cannot be used to study the fine structure in the polymer spectrum due to water vapor interference. The present invention obtains the absorbance spectrum of a pure sample without water vapor interference and then performs a second derivative thereof, thereby observing the fine structure in the spectrum.

[0132] According to the present invention, the second-order derivative can be obtained by formula X. Specifically, assuming that A is a pure sample spectrum obtained without water vapor interference, there are x data points on it, and assuming that the second-order derivative of A is B, then there is formula X:

[0133] B(x)=A(x-1)+A(x+1)-2×A(x) Formula X.

[0134] For example, for the fifth data point in B, x=5, and B(x) is the fourth data point of A plus the sixth data point of A minus twice the fifth data point of A.

[0135] The second aspect of the present invention aims to provide an application of the method described in the first aspect for removing interference from water vapor peaks in infrared spectra.

[0136] According to the present invention, the application object is not limited to any substance, and not only can a pure sample spectrum without water vapor interference be obtained, but also fine structures in the spectrum can be observed.

[0137] Examples and Comparative Examples

[0138] The present invention is further described below by specific examples. However, these examples are merely exemplary and do not constitute any limitation to the protection scope of the present invention.

[0139] The spectra of the following examples were collected on a Thermo-Fischer Nicolet-6700 FTIR spectrometer. All spectra were measured at 2 cm-1 The scans were recorded at a resolution of 1000 x 200 and a total of 128 scans were added.

[0140] Example 1

[0141] This example uses ethanol as a research object, and the ethanol is AR grade. The infrared absorbance spectrum without water vapor interference is obtained according to the REACT-FILES method of the present invention.

[0142] (1) Obtain the FTIR single-beam spectrum of gas-phase ethanol in the laboratory, such as Figure 1 As shown in A, by taking the negative logarithm of the spectrum library spectrum and the gas phase ethanol spectrum according to formula I and taking the first-order derivative according to formula II, according to formula III, the peak positions are located at 3850cm -1 、3851cm -1 、3852cm -1 、3853cm -1 、3854cm -1 、3855cm -1 The phase angle is calculated from the spectral data at , and the two smallest phase angle values ​​are 0.0823° and 0.0712°, which are the two single-beam background spectra selected from the spectral library that are closest to the sample peak position offset, respectively. Figure 1 As shown in Figures 1B and 2B.

[0143] (2) Take the negative logarithm of the above three single-beam spectra according to formula I, and complete the two-dimensional asynchronous correlation spectrum. The two-dimensional asynchronous correlation spectrum is as follows: Figure 2 As shown, the top and right sides of the figure are single-beam spectra of sample gas-phase ethanol.

[0144] (3) Step (3) is exactly the same as step (2), except that the negative logarithm of the three single-beam spectra is taken according to formula I, the first-order derivative is taken according to formula II, and the two-dimensional asynchronous correlation spectrum is completed according to formulas IV to VII. The two-dimensional asynchronous correlation spectrum is as follows: Figure 3 As shown, the top and right sides of the figure are the first derivative spectra of the negative logarithm of the single-beam spectrum of gas-phase ethanol of the sample.

[0145] The above data processing computer simulation and the generation of two-dimensional asynchronous correlation spectrum are completed by using MATLAB software.

[0146] (4) Gaseous ethanol spectrum in the Aldrich spectral library (corresponding to Figure 4 Curve A), sample ethanol absorbance spectrum (corresponding to Figure 4 The middle curve B), the absorbance spectrum obtained by the negative logarithmic single beam two-dimensional spectrum intercept of the sample ethanol (corresponding to Figure 4 The middle curve C), the intersection position is 3906.110cm -1, the sample ethanol spectrum is obtained by taking the negative logarithm and then taking the derivative to construct a two-dimensional spectrum section, and the absorbance spectrum obtained by integration and reduction (corresponding to Figure 4 The middle curve D), where the intercept position is 3786.545cm -1 At , the above spectra are summarized as follows Figure 4 shown.

[0147] Combination Figures 2-3 The two-dimensional asynchronous correlation spectrum and the corresponding spectra on the top and right of the figure show that:

[0148] Water vapor at a wavelength of 4000cm -1 ~3800cm -1 However, in the two-dimensional asynchronous spectrum, there is an absorption peak near 4000cm -1 ~3800cm -1 There is no cross peak, that is, there is a systematic peak missing, as shown by the red box in the figure.

[0149] Combination Figures 2 to 4 , we can see that:

[0150] The absorbance of gaseous ethanol (corresponding to Figure 4 The middle curve B) has very serious interference from moisture;

[0151] The spectrum obtained by the method of step (2) is the negative logarithmic single beam two-dimensional spectrum of the sample ethanol (corresponding to Figure 4 Curve C) can effectively alleviate the interference of moisture on the obtained spectrum, but there is still residual moisture interference. Figure 4 Curve C is marked with a rectangle because there is still a non-zero baseline in the negative logarithm of the single-beam spectrum, and the non-zero baseline destroys the fragile orthogonality;

[0152] The spectrum obtained by the method of step (3) is the REACT-FILES method. Specifically, the negative logarithmic derivative of the sample ethanol single beam spectrum and the background single beam spectrum is obtained to construct a two-dimensional spectrum cut line, and the regenerated spectrum (corresponding to Figure 4 The middle curve D) can eliminate the baseline at 3676cm -1 The Q branch of the OH stretching peak can be clearly seen at 2 Bending peak (at Figure 4 The inset of curve A shows a broad peak. Figure 4 The illustration of D shows a double peak (including P branch and R branch). It can be seen that the spectral curve obtained by the REACT-FILES method has a higher resolution, removes the water vapor peak more thoroughly, and has a better effect. The REACT-FILES method avoids the interference introduced by manual baseline correction, thereby obtaining a gas-phase ethanol regeneration spectrum without water vapor interference.

[0153] Comparative Example 2

[0154] In order to compare the regeneration spectrum of gas-phase ethanol without water vapor interference obtained by the REACT-FILE method of the present invention to the absorbance spectrum of gas-phase ethanol obtained by the difference spectrum subtraction method, wherein the gas-phase ethanol is exactly the same as the ethanol in Example 1, the specific steps are as follows:

[0155] The FTIR single beam spectrum of gas phase ethanol and the matching single beam background spectrum are exactly the same as those in Example 1, and are represented as B ethanol (x), B Ba (corresponding to the single beam background spectrum 1B in Example 1) and B Bb (corresponding to the single beam background spectrum 2B in Example 1), the negative logarithm is taken to obtain the absorbance A: At this time, A is the absorbance spectrum containing water vapor and sample ethanol, and the negative logarithm is taken to make the absorbance B: At this time, B is the absorbance spectrum containing only water vapor, and the difference spectrum C is made: C=A+k·B, where k is the subtraction factor.

[0156] The subtraction factor range is set to -30 to 30, and a subtraction spectrum is made for every 0.0001 value change. For example, the subtraction factor may be -30, -29.9999, -29.9998, -29.9997, ..., 0, ... 29.9998, 29.9999, 30, from negative to positive, so that C = A + k·B can reach the cleanest state of water vapor reduction; a total of 600,001 subtraction spectra are made, and finally, one at 4000cm is selected from the 600,001 candidate samples. -1 ~3800cm -1 The spectrum with the lowest standard deviation between the two is the cleanest spectrum after water vapor subtraction. Then the spectrum is normalized so that the 1066cm -1 The intensity of the peak at 1066 cm is 1.0 because -1 The peak is the strongest peak in the FTIR spectrum of gaseous ethanol. Figure 5 The inset of curve A shows the -1 ~3700cm -1 The spectrum between 1700cm -1 ~1500cm -1 Residual interference of water can still be observed in the spectra between .

[0157] The absorbance spectrum of gas phase ethanol without water vapor interference obtained in Example 1 is as follows Figure 5 As shown in the curve B, Figure 5 In the inset of curve B, the water content at 1700 cm-1 Up to 1500cm -1 The residual interference in the spectral region between is significantly smaller than that obtained using difference spectrum subtraction.

[0158] Figure 5 It can be clearly seen that the fluctuation amplitude of the spectrum obtained using the subtraction method is greater than that of the spectrum obtained using the REACT-FILES method.

[0159] Calculate 1710cm -1 ~1550cm -1 and 3900cm -1 ~3750cm -1 The standard deviation (Std value) of the two spectral intensities in the spectral region between is shown in Table 1:

[0160] method <![CDATA[1710cm -1 ~1550cm -1 Std]]> <![CDATA[3900cm -1 ~3750cm -1 Std]]> Difference Spectrum Subtraction <![CDATA[1.62×10 -3 ]]> <![CDATA[1.50×10 -3 ]]> REACT-FILES <![CDATA[1.54×10 -3 ]]> <![CDATA[8.0×10 -3 ]]>

[0161] From the calculation results in Table 1, it can be seen that the spectral area of ​​the Std value of the spectrum generated by the REACT-FILES method is lower than that of the spectrum generated by the subtraction method, which further confirms that the fluctuation amplitude of the spectrum obtained by the REACT-FILES method is lower.

[0162] Example 3

[0163] For polyethylene (PE), CH 2 The bending zone is located at 1500cm -1 ~1400cm -1 The spectrum of polyethylene (PE) obtained using the REACT-FILES method is shown in the figure Figure 6 As shown, CH 2 The bending zone is shown as a rectangular box in the figure, and the lower part of the inset shows the bending zone at 1550cm -1 ~1400cm -1 In this case, the interference of moisture can be completely eliminated from the generated spectrum and the corresponding second-order derivative spectrum, where the two-dimensional asynchronous correlation spectrum cut-off position is 3937.93 cm -1 In the inset, the orthorhombic crystalline form of polyethylene can be clearly observed at 1473 cm -1 and 1463cm -1 For comparison, the PE peak at 1550cm was obtained by conventional method. -1 ~1400cm -1 The second derivative spectrum of the FTIR spectrum between the two, as shown in the upper part of the inset, shows that the interference of water vapor completely masks the CH 2 Fine spectral structure of the bending band.

[0164] Example 4

[0165] In ethylene-vinyl acetate copolymer (EVA), it can be -1 ~1700cm -1 The carbonyl band of the ester group is observed in the spectral region between . The EVA spectrum obtained using the REACT-FILES method is shown in Figure 7 The carbonyl band of the ester group is shown in the rectangular box, and the lower part of the inset shows the 1800cm -1 ~1660cm -1 The corresponding second-order derivative spectra in the spectral region between 2 and 4 clearly show the fine spectral structure of the carbonyl band, where the two-dimensional asynchronous correlation spectrum cut-off position is 3921.538 cm -1 For comparison, the conventional method was used to obtain EVA at 1800cm -1 ~1660cm -1 The second derivative spectrum of the FTIR spectrum, as shown in the upper part of the inset, shows that the presence of water makes it difficult to identify the fine spectral structure of the carbonyl band.

[0166] The present invention is described in detail above in combination with preferred embodiments and exemplary examples. However, it should be noted that these specific embodiments are only illustrative explanations of the present invention and do not constitute any limitation on the protection scope of the present invention. Without exceeding the spirit and protection scope of the present invention, various improvements, equivalent substitutions or modifications may be made to the technical content of the present invention and its embodiments, which all fall within the protection scope of the present invention. The protection scope of the present invention shall be subject to the attached claims.

Claims

1. A method for removing interference from water vapor peaks in infrared spectra, characterized in that: The method comprises: Step 1, select two single-beam background spectra from the spectrum library; Step 2, constructing a two-dimensional asynchronous correlation spectrum according to the single-beam sample spectrum and the single-beam background spectrum selected in step 1; Step 3, processing the two-dimensional asynchronous correlation spectrum obtained in step 2 to obtain an infrared spectrum of the sample without water vapor interference, In step 1, the single beam background spectrum is recorded and stored in a spectrum library, which includes single beam background spectra from different seasons, different time periods, and different room temperature conditions, including single beam background spectra of the helium-neon laser cavity of the infrared spectrometer under different temperature conditions, The step 2 comprises the following steps: Step 2-1, converting the first-order derivative of the negative logarithm of the single-beam sample spectrum and the two selected single-beam background spectra into a dynamic spectrum; Step 2-2, constructing the dynamic spectrum obtained in step 2-1 into a two-dimensional asynchronous correlation spectrum, In step 3, the processing is to obtain a pure sample spectrum without water vapor interference by taking a line cut at a systematic missing peak in the two-dimensional asynchronous correlation spectrum and integrating the line cut.

2. The method according to claim 1, characterized in that: In step 1, the single beam background spectrum selected from the spectrum library is determined according to the phase angle between the single beam background spectrum and the single beam sample.

3. The method according to claim 1 or 2, characterized in that: Step 1 includes the following sub-steps: Step 1-1, taking negative logarithms and derivatives of the single-beam sample spectrum and the single-beam background spectrum in the spectral library; Step 1-2, using the derivative data obtained in step 1-1 to obtain the phase angle between the single beam sample spectrum and the single beam background spectrum; Step 1-3, selecting two single-beam background spectra with the lowest phase angles according to the phase angles obtained in step 1-2.

4. The method according to claim 3, characterized in that In step 1-1, the negative logarithm formula is as shown in Formula I: L S (x)=-log 10 (B S (x)) L B (x)= -log 10 (B B (x)) Formula I; Where: x is the wave number of infrared light; B S (x) is the spectral intensity of the single beam sample spectrum Bs at the x wave number; B B (x) is the single beam background spectrum B B Spectral intensity at wave number x; L S (x) is the negative logarithm of the spectral intensity of the single beam sample spectrum Bs at the x wave number; L B (x) is the single beam background spectrum B B Negative logarithm of the spectral intensity at wavenumber x.

5. The method according to claim 3, characterized in that: In step 1-1, the derivative formula is shown in Formula II: Where: L' S (x) is the first derivative of the negative logarithm of the spectral intensity of the single-beam sample spectrum Bs at the x wavenumber; L' B (x) is the single beam background spectrum B B The first derivative of the negative logarithm of the spectral intensity at wavenumber x.

6. The method according to claim 3, characterized in that In step 1-2, the phase angle calculation formula is shown in Formula III: Where: θ is the single beam sample spectrum L' S (x i ) and single beam background spectrum L' B (x i ) between the phase angles; x i 4000cm -1 ~3750cm -1 Spectral data points between; L' S (x i ) is the single beam sample spectrum Bs at x i The first derivative of the negative logarithm of the spectral intensity at the spectral data point; L' B (x i ) is the single beam background spectrum B B In x i The first derivative of the negative logarithm of the spectral intensity at the spectral data point.

7. The method according to claim 6, characterized in that x i 3855cm -1 ~3850cm -1 spectral data points between .

8. Use of the method according to any one of claims 1 to 7 for removing interference from water vapor peaks in infrared spectra.

Images