Method for determining polyalphaolefin content based on spectroscopy

Abstract

The application relates to the field of spectrum technology and discloses a method for determining the content of poly-alpha-olefins based on spectrum. The method comprises the following steps: collecting a first spectrum and a second spectrum of a sample to obtain a spectrum matrix; constructing a function model P1 of the spectrum matrix of the first spectrum and the content of poly-alpha-olefins, and substituting the spectrum matrix of the first spectrum into P1 to obtain Y pred , T1 and y r1 ; orthogonizing the spectrum matrix of the second spectrum and T1 to obtain X orth ; constructing a model P2 of X orth and y r1 ; obtaining the spectrum matrix of the first spectrum and the spectrum matrix of the second spectrum of a to-be-detected liquid, substituting the spectrum matrix of the first spectrum of the to-be-detected liquid into P1 to obtain T1 M and Q1 M ; substituting the spectrum matrix of the second spectrum of the to-be-detected liquid into P2 to obtain a model score and a load, and substituting into a calculation model to obtain the content of poly-alpha-olefins. The method is simple in operation, fast in speed and accurate in result.

Description

Technical Field

[0001] This invention relates to the field of spectroscopy, and more specifically to a method for determining the content of polyalphaolefins based on spectroscopy. Background Technology

[0002] Polyalphaolefin (PAO) synthetic oils are olefin oligomers obtained by polymerizing monomeric alpha-olefins under the action of a catalyst and then refining them through hydrogenation. Their molecular structures are regular and uniform, exhibiting a comb-like or dendritic appearance. Among PAOs, PAO base oils produced using 1-decene as a raw material have the best performance. The higher the polyalphaolefin content in PAO, the better the product's properties, including thermal stability, thermal oxidation stability, and color.

[0003] Currently, methods for determining the polyalphaolefin (PAO) content in polyolefins (PAOs) mainly include high-performance liquid chromatography (HPLC), nuclear magnetic resonance (NMR) analysis, and gel permeation chromatography (GPC). However, these methods involve cumbersome manual operations, require highly skilled analysts, are significantly affected by manual operation in terms of accuracy, are slow, and the environmental impact of the reagents used cannot be ignored, making them less than ideal for industrial production. Therefore, there is still a need to develop a rapid, simple, and accurate method for predicting the PAO content. Summary of the Invention

[0004] The purpose of this invention is to overcome the above-mentioned problems in the prior art and provide a method for determining the content of polyalphaolefins based on spectroscopy. This method is simple to operate, fast, accurate, and has good repeatability.

[0005] To achieve the above objectives, the present invention provides a method for determining the content of polyα-olefins based on spectroscopy, the method comprising:

[0006] (1) Obtain multiple standard polyα-olefin samples with known and different polyα-olefin contents;

[0007] (2) Collect the first and second spectra of the above standard polyα-olefin samples respectively;

[0008] Select the characteristic spectral bands of the above spectra respectively, and obtain the corresponding spectral matrices;

[0009] (3) Using partial least squares method, construct a functional model P1 of the spectral matrix of the first spectrum and the polyalphaolefin content of the standard polyalphaolefin sample, and substitute the spectral matrix of the first spectrum into the functional model P1 to obtain the calculated value Y of the first polyalphaolefin content. pred Model score T1, true value of polyα-olefin content and Y pred The model residual matrix y r1 ;

[0010] (4) Orthogonalize the spectral matrix of the second spectrum with the model score T1 to obtain the orthogonalized matrix X. orth ;

[0011] (5) Construct the orthogonal matrix X using the partial least squares method. orth With y r1 Function model P2;

[0012] (6) Take a polyα-olefin sample with unknown polyα-olefin content as the test solution, collect the spectrum according to the method in step (2), obtain the spectral matrix of the first spectrum of the test solution and the spectral matrix of the second spectrum of the test solution, and substitute the spectral matrix of the first spectrum of the test solution into the function model P1 to obtain the model score T1. M and load Q1 M ;

[0013] (7) Substitute the spectral matrix of the second spectrum of the test liquid into the function model P2 to obtain the model score T2. M and load Q2 M Substitute into the calculation model Y=T1 M ×Q1 M +T2 M ×Q2 M The content Y of poly-α-olefin in the test solution was obtained.

[0014] The technical solution of this invention is simple to operate, fast, accurate, and has good repeatability. It integrates different spectra, allowing for complementary information between the spectra, resulting in comprehensive and accurate results suitable for practical industrial production. Detailed Implementation

[0015] The endpoints and any values ​​of the ranges disclosed herein are not limited to the precise ranges or values, and these ranges or values ​​should be understood to include values ​​close to these ranges or values. For numerical ranges, the endpoint values ​​of the various ranges, the endpoint values ​​of the various ranges and individual point values, and individual point values ​​can be combined with each other to obtain one or more new numerical ranges, which should be considered as specifically disclosed herein.

[0016] This invention provides a method for determining the content of polyα-olefins based on spectroscopy, the method comprising:

[0017] (1) Obtain multiple standard polyα-olefin samples with known and different polyα-olefin contents;

[0018] (2) Collect the first spectrum and the second spectrum (the first spectrum and the second spectrum are different) for the above standard polyα-olefin samples;

[0019] Select the characteristic spectral bands of the above spectra respectively, and obtain the corresponding spectral matrices;

[0020] (3) Using partial least squares method, construct a functional model P1 of the spectral matrix of the first spectrum and the polyalphaolefin content of the standard polyalphaolefin sample, and substitute the spectral matrix of the first spectrum into the functional model P1 to obtain the calculated value Y of the first polyalphaolefin content. pred Model score T1, true value of polyα-olefin content and Y pred The model residual matrix y r1 ;

[0021] (4) Orthogonalize the spectral matrix of the second spectrum with the model score T1 to obtain the orthogonalized matrix X. orth ;

[0022] (5) Construct the orthogonal matrix X using the partial least squares method. orth With y r1 Function model P2;

[0023] (6) Take a polyα-olefin sample with unknown polyα-olefin content as the test solution, collect the spectrum according to the method in step (2), obtain the spectral matrix of the first spectrum of the test solution and the spectral matrix of the second spectrum of the test solution, and substitute the spectral matrix of the first spectrum of the test solution into the function model P1 to obtain the model score T1. M and load Q1 M ;

[0024] (7) Substitute the spectral matrix of the second spectrum of the test liquid into the function model P2 to obtain the model score T2. M and load Q2 M Substitute into the calculation model Y=T1 M ×Q1 M +T2 M ×Q2 M The content Y of poly-α-olefin in the test solution was obtained.

[0025] It's understandable that different spectra will produce different ordinates for the spectral data. For example, the ordinate of infrared and ultraviolet spectra is generally absorbance, resulting in an absorbance matrix. In contrast, the ordinate of Raman spectra is generally intensity, resulting in an intensity matrix. Taking the first spectrum as an example, one sample corresponds to a 1×N spectral matrix, where N is the wavelength; multiple samples correspond to an M×N spectral matrix, where M is the number of samples and N is the wavelength.

[0026] The inventors of this invention discovered in their research that, according to the above-described method, for standard polyalphaolefins, by first utilizing the first spectrum and then fusing the data obtained from the first spectrum with the second spectrum, the two spectral information are processed in series, and the final result is given jointly by the two spectra, which can further ensure the accuracy of the result. The method of this invention can fully integrate complementary information from different spectra, achieving complementary advantages between spectral analysis techniques, resulting in more accurate results. Using the method of this invention, the polyalphaolefin content of polyalphaolefin samples with unknown content can be quickly and accurately determined. Compared with existing commonly used methods for determining the polyalphaolefin content in polyalphaolefin samples, such as high-performance liquid chromatography, nuclear magnetic resonance analysis, and gel permeation chromatography, the method of this invention provides accurate results, avoids complex sample processing procedures, and is simpler and faster to operate.

[0027] According to the present invention, preferably, in step (1), the number of standard poly-α-olefin samples is ≥50, and more preferably 60-200.

[0028] Generally, the sample collection should ensure that the sample is evenly distributed within the possible range. This invention does not impose any particular limitation on the polyalphaolefin content in the standard polyalphaolefin sample. However, preferably, the polyalphaolefin content in the standard polyalphaolefin sample satisfies the following: a mass concentration greater than 1 wt%, preferably 50-99.8 wt% (for example, it can be 50 wt%, 55 wt%, 60 wt%, 65 wt%, 70 wt%, 75 wt%, 80 wt%, 85 wt%, 90 wt%, 95 wt%, 98 wt%, 99 wt%, 99.8 wt%, and any two of the above values ​​within a range).

[0029] This invention does not impose any particular restrictions on the spectral acquisition conditions, and can use conditions commonly used in the field. Preferably, the acquisition temperature is 15-50℃, and more preferably 23-27℃.

[0030] According to the present invention, preferably, the resolution of the acquisition is 2-16cm. -1 More preferably 5-12cm -1 (For example, it can be 5cm) -1 6cm -1 7cm -1 8cm -1 9cm -1 10cm -1 11cm -1 12cm -1 And the range formed by any two of the above values, and the values ​​within that range).

[0031] According to the present invention, preferably, the first spectrum and the second spectrum are each independently selected from one of mid-infrared spectroscopy, Raman spectroscopy, near-infrared spectroscopy, laser-induced breakdown spectroscopy, and ultraviolet spectroscopy, and the first spectrum and the second spectrum are different. It is understood that the first spectrum and the second spectrum are different in order to fuse complementary information from different spectra and make the results of the computational model more accurate.

[0032] According to a preferred embodiment of the present invention, the first spectrum and the second spectrum are selected from mid-infrared spectroscopy and Raman spectroscopy. The first spectrum can be mid-infrared spectroscopy, and the second spectrum can be Raman spectroscopy; or, the first spectrum can be Raman spectroscopy, and the second spectrum can be mid-infrared spectroscopy. However, according to a particularly preferred embodiment of the present invention, the first spectrum is mid-infrared spectroscopy, and the second spectrum is Raman spectroscopy. The inventors of the present invention further discovered in their research that, for polyα-olefin samples, using mid-infrared spectroscopy and Raman spectroscopy results in better accuracy compared to other spectral methods.

[0033] According to the present invention, preferably, in step (2), the wavenumber range for acquiring the mid-infrared spectrum is 650-4000 cm⁻¹. -1 When collecting mid-infrared spectra, an Agilent 4500tFTIR Fourier transform infrared spectrometer can be used. Take a drop of crude oil sample with a pipette and add it to the sample cell. Use air as a reference for spectral scanning. The number of sample scans can be 16-128.

[0034] According to the present invention, preferably, the wavenumber range for Raman spectroscopy acquisition is 100-2500 cm⁻¹. -1 When collecting Raman spectra, a handheld Raman spectrometer (Thermo Fisher Scientific, model TruScanRM) can be used, with an emission wavelength of 100-3000 nm. The sample is placed in a vial for measurement, and the laser power can be 1-450 mW. The number of sample scans can be 16-128.

[0035] According to the present invention, preferably, before selecting the characteristic spectral bands, the method further includes: preprocessing the spectrum to eliminate redundant information and / or noise in the spectrum. It is understood that preprocessing can eliminate baseline drift, spectral overlap, and noise in the spectrum, thereby further improving the accuracy of the results.

[0036] According to the present invention, preferably, the preprocessing method is selected from at least one of second-order differential, multivariate scattering correction, first-order differential and standard normal variable transformation; more preferably, second-order differential is used for preprocessing of mid-infrared spectra and multivariate scattering correction is used for preprocessing of Raman spectra.

[0037] The window width for the second derivative can be 20-32 (e.g., 20, 21, 24, 25, 27, 30, 32).

[0038] According to the present invention, preferably, the wavenumber range of the characteristic spectral band of the mid-infrared spectrum is 650-4000 cm⁻¹. -1 More preferably, 1603-1663cm -1 and 650-980cm -1 .

[0039] Preferably, the wavenumber range of the characteristic spectral band of the Raman spectrum is 100-2500 cm⁻¹. -1 More preferably, 1018-1755cm -1 .

[0040] The inventors of this invention further discovered in their research that by selecting the above-mentioned wavelength bands, information about carbon molecules related to poly-α-olefins in the spectrum can be fully obtained, which can further ensure the accuracy of the results.

[0041] It is understandable that in step (3), taking the spectral matrix corresponding to the characteristic spectral band of the first spectrum as an example, multiple standard poly-α-olefin samples form an “M×N matrix”, where M is the number of samples and N is the wavelength.

[0042] Preferably, after obtaining the corresponding spectral matrix, the method further includes: normalizing the spectral matrix. It is understood that the purpose of normalization is to scale the spectral intensity (normalizing the highest value, while other intensities are represented as decimals less than 1), avoiding the loss of detail in spectral information due to differences in absorbance or spectral intensity.

[0043] In steps (3), (6), and (7), the model scores and loads are parameters derived from the partial least squares method. It is understandable that the loads can also be obtained in step (3); in step (6), after substituting the spectral matrix into the function model P1, the calculated value of model P1 can also be obtained simultaneously; in step (7), after substituting the spectral matrix into the function model P2, the calculated value of model P2 can also be obtained simultaneously.

[0044] According to the present invention, preferably, in step (4), the orthogonal method is selected from Schmidt orthogonalization and / or normalized orthogonalization, more preferably Schmidt orthogonalization. By orthogonalization, the information that is repeated between different elimination spectra can be fully extracted, while the complementary information is fully extracted, thus fusing the different information of different spectra, thereby making the results more accurate.

[0045] For example, in step (6), a polyα-olefin sample with an unknown polyα-olefin content is taken, and a spectral matrix corresponding to the characteristic spectral band of the first spectrum and a spectral matrix corresponding to the characteristic spectral band of the second spectrum are formed; the above matrices are respectively "1×N matrices", where N is the wavelength.

[0046] The function model P2 is related to the orthogonalization matrix X. orth With y r1 The model. In step (7), when the spectral matrix of the second spectrum of the test liquid is substituted into the function model P2, the spectral matrix of the second spectrum corresponds to X. orth Substitute.

[0047] According to the present invention, preferably, the sample is a polyalphaolefin synthetic oil. After obtaining the polyalphaolefin synthetic oil, the polyalphaolefin content can be determined by chromatography, for example, using the ASTM D6352 method. The polyalphaolefin content is determined based on the boiling point. Polyalphaolefins have high boiling points; components with boiling points exceeding 300°C are considered to be polyalphaolefins.

[0048] Understandably, polyalphaolefin synthetic oils produced from 1-decene generally exhibit the best performance. Therefore, the polyalphaolefin in polyalphaolefin synthetic oils preferably refers to hydrogenated poly1-decene. Polyalphaolefin synthetic oils typically also contain impurities such as 1-dodecene, some linear alpha-olefins, β-olefins, and small amounts of aromatics and alkanes.

[0049] The present invention will be described in detail below through embodiments. In the following embodiments,

[0050] The mid-infrared spectrum of the sample was acquired using a Fourier transform infrared spectrometer (Agilent 4500t FTIR).

[0051] Raman spectra of the samples were acquired using a handheld Raman spectrometer (TruScan RM, ThermoFisher).

[0052] In step (1) of the following embodiments,

[0053] The content of polyalphaolefins is determined by gas chromatography according to the ASTM D6352 method, based on the boiling point. Polyalphaolefins have a high boiling point; components with a boiling point exceeding 300°C are considered to be polyalphaolefins.

[0054] The operations following the acquisition of the spectrum are performed in MATLAB.

[0055] Example 1

[0056] This invention provides a method for determining the content of polyα-olefins based on spectroscopy.

[0057] (1) 100 polyalphaolefin synthetic oil (PAO) samples were obtained and the polyalphaolefin content (distributed in the range of 69.23-99.13 wt%) was determined by gas chromatography.

[0058] (2) For the above PAO samples, mid-infrared spectra were collected: a drop of crude oil sample was added to the sample cell using a pipette, and a spectral scan was performed using air as a reference, in the range of 650-4000 cm⁻¹. -1 Within an area of ​​8cm -1 The spectrum was acquired at a high resolution, with 64 sample scans and an acquisition temperature of 24℃.

[0059] Raman spectra of PAO samples were acquired, with the emitted laser wavelength at 785 nm. The PAO samples were measured using vials with a laser power of 250 mW, within the range of 100-2500 cm⁻¹. -1 Within an area of ​​8cm -1 The spectrum was acquired at a high resolution, with 64 sample scans and an acquisition temperature of 24℃.

[0060] The acquired mid-infrared spectra were processed using a second-order derivative with a window width of 27 points, and the wavenumber range was selected as 1603-1663 cm⁻¹. -1 and 650-980cm -1 The region is taken as the characteristic spectral band, and the corresponding absorbance matrix X is obtained. MIR ;

[0061] The acquired Raman spectra were subjected to multivariate scattering correction, with the wavenumber range being 10¹⁸–1755 cm⁻¹. -1 The region is taken as the characteristic spectral band, and the corresponding spectral intensity matrix X is obtained. Raman ;

[0062] The absorbance and spectral intensity of the above characteristic spectral bands are normalized respectively (the highest value of absorbance is normalized, and other values ​​are correspondingly decimals less than 1; or the highest value of spectral intensity is normalized, and other values ​​are correspondingly decimals less than 1).

[0063] (3) Construct the absorbance matrix X using the partial least squares method. MIR The function model P1 of the poly-α-olefin content of the PAO sample, and the absorbance matrix X of the mid-infrared spectrum. MIR Substituting into P1, we obtain the calculated value Y. pred And simultaneously obtain the model score T1, the true value of polyα-olefin content, and Y. pred The model residual matrix y r1 .

[0064] (4) The spectral intensity matrix XRaman Performing a Schmitt orthogonal analysis with the model score T1 yields the orthogonalized matrix X. orth ;

[0065] (5) Construct the orthogonal matrix X using the partial least squares method. orth With y r1 Function model P2;

[0066] (6) Take 24 polyalphaolefin synthetic oil samples with unknown polyalphaolefin content as a validation set, and determine the polyalphaolefin content in them by gas chromatography (see the measured values ​​in Table 1, which are distributed in the range of 63.98-98.76 wt%).

[0067] For each sample to be tested, perform the following processing steps up to (7):

[0068] Following step (2), the mid-infrared spectrum, Raman spectrum, and their corresponding characteristic spectral bands, as well as the corresponding absorbance matrix and spectral intensity matrix (which were also normalized in the same way) were obtained.

[0069] Substituting the absorbance matrix of the test liquid into the function model P1, the calculated value Y is obtained. pred M And simultaneously obtain the model score T1 M and load Q1 M ;

[0070] (7) Substitute the spectral intensity matrix of the test liquid into the function model P2 to obtain the calculated value Y. rpred M Model score T2 M and load Q2 M Then the polyα-olefin content Y in the test solution is Y = T1 M ×Q1 M +T2 M ×Q2 M .

[0071] The measured values ​​obtained by chromatography for each test solution in the validation set, the Y values ​​calculated by the method of this invention (i.e., the calculated values ​​in Table 1), and the error between the two (calculated value minus measured value) are shown in Table 1.

[0072] The performance of the method of the present invention is evaluated using SEP (standard deviation of prediction) and RPD (ratio of standard deviation to root mean square error).

[0073]

[0074]

[0075] In the above formula, i represents the i-th sample, i = 1, 2, ..., m, where m is the total number of samples in the validation set (24), and y i,actual Let y be the gas chromatographic determination value of the i-th sample. i,predicted SD is the calculated value of the i-th sample obtained by the method of this invention. V To verify the standard deviation of polyalphaolefin content in all samples of the validation set, SD V The calculation formula is as follows:

[0076]

[0077] Where i = 1, 2, ..., N, N is the number of samples in the validation set (24), X i This represents the polyalphaolefin content (determined by gas chromatography) of the i-th sample. The average polyalphaolefin content (determined by gas chromatography) of all samples in the validation set.

[0078] The SEP was found to be 0.34 and RPD to be 24.71, indicating that the method of the present invention has high accuracy. Furthermore, the inventors of the present invention also discovered that the number of principal factors for function model P1 is 4, and the number of principal factors for function model P2 is 5. It is understood that, according to partial least squares, the number of principal factors represents the complexity of the model; a fewer principal factors indicate that the model is closer to linear, and a more linear model has stronger generalization ability, i.e., better universality. This also demonstrates that the method of the present invention has high accuracy.

[0079] Table 1

[0080]

[0081]

[0082] It is evident that the method provided by this invention has a smaller error and higher accuracy.

[0083] Example 2

[0084] This is used to illustrate the repeatability of the method provided by the present invention.

[0085] Take a sample with unknown polyalphaolefin content and determine its polyalphaolefin content according to the chromatographic method in step (1) of Example 1 (i.e., the measured values ​​in Table 2).

[0086] For this sample, six parallel spectral acquisitions were performed. After each parallel spectral acquisition, mid-infrared and Raman spectra were acquired according to step (2) in Example 1. The characteristic spectral bands were selected according to step (2), and the corresponding absorbance matrix and spectral intensity matrix were obtained (normalized as well). The absorbance matrix of the test liquid was substituted into the function model P1 to obtain the calculated value Y. predM ', and simultaneously obtain the model score T1 M 'and load Q1 M Substitute the spectral intensity matrix of the test liquid into the function model P2 to obtain the calculated value Y. rpred M ', Model score T2 M’ and load Q2 M’ Then the polyα-olefin content Y in the test solution is Y = T1 M '×Q1 M '+T2 M '×Q2 M '.

[0087] For this sample, six parallel results (calculated values) were obtained. The error for each result is the difference between the calculated value and the gas chromatography measurement value. The repeatability of the model was evaluated using RSD (Repeat Score Determination), with a smaller value indicating better repeatability. The formula for calculating RSD is as follows:

[0088]

[0089] Where S represents the standard deviation of the calculated value, The x represents the average of the calculated values ​​of the sample, i = 1, 2, ..., n, where n represents the number of samples. i This represents the calculated value of the i-th sample.

[0090] The results are shown in Table 2.

[0091] Table 2

[0092]

[0093]

[0094] It is evident that the method provided by this invention has good repeatability.

[0095] Comparative Example 1

[0096] For the 100 samples in step 1 of Example 1, follow steps (1)-(2) of Example 1, and then construct the spectral intensity matrix X using the partial least squares method. Raman The function model P3 is related to the polyalphaolefin content of the PAO sample.

[0097] Then, the validation set (24 samples) of Example 1 was taken, and the absorbance matrix of each test liquid was substituted into the function model P1 in Example 1 (to obtain the calculated value using P1 alone), and the spectral intensity matrix of each test liquid was substituted into the function model P3 (to obtain the calculated value using P3 alone).

[0098] Table 3 shows the cases where SEP and RPD are used individually with P1 or P3. The calculation formulas for SEP and RPD are as follows.

[0099]

[0100]

[0101] Where i represents the i-th sample, i = 1, 2, ..., m, and m is the total number of samples in the validation set (24). i,actual Let y be the gas chromatographic determination value of the i-th sample. i,predicted For the calculated value of the i-th sample obtained using P1 or P3 alone, SD V To verify the standard deviation of polyalphaolefin content in all samples of the validation set, SD V The calculation formula is as follows:

[0102]

[0103] Where i = 1, 2, ..., N, N is the number of samples in the validation set (24), X i This represents the polyalphaolefin content (determined by gas chromatography) of the i-th sample. The average polyalphaolefin content (determined by gas chromatography) of all samples in the validation set.

[0104] Table 3

[0105] SEP RPD <![CDATA[P1]]> 0.54 15.78 <![CDATA[P3]]> 0.60 14.18 Example 1 0.34 24.71

[0106] The smaller the SEP and the larger the RPD, the better the model accuracy. It can be seen that the error using the scheme provided by this invention is significantly smaller than the error of a model directly constructed using only mid-infrared and Raman spectroscopy.

[0107] The preferred embodiments of the present invention have been described in detail above; however, the present invention is not limited thereto. Within the scope of the inventive concept, various simple modifications can be made to the technical solutions of the present invention, including combinations of various technical features in any other suitable manner. These simple modifications and combinations should also be considered as the content disclosed in the present invention and are all within the protection scope of the present invention.

Claims

1. A method for determining the content of polyα-olefins based on spectroscopy, characterized in that, The method includes: (1) Obtain multiple standard polyα-olefin samples with known and different polyα-olefin contents; (2) Collect the first and second spectra of the above standard poly-α-olefin samples respectively; Select the characteristic spectral bands of the above spectra respectively, and obtain the corresponding spectral matrices; (3) Using partial least squares method, construct a functional model P1 of the spectral matrix of the first spectrum and the polyalphaolefin content of the standard polyalphaolefin sample, and substitute the spectral matrix of the first spectrum into the functional model P1 to obtain the calculated value Y of the first polyalphaolefin content. pred Model score T1, true value of polyα-olefin content and Y pred The model residual matrix y r1 ; (4) Orthogonalize the spectral matrix of the second spectrum with the model score T1 to obtain the orthogonalized matrix X. orth ; (5) Construct the orthogonal matrix X using the partial least squares method. orth With y r1 Function model P2; (6) Take a polyα-olefin sample with unknown polyα-olefin content as the test solution, collect the spectrum according to the method in step (2), obtain the spectral matrix of the first spectrum of the test solution and the spectral matrix of the second spectrum of the test solution, and substitute the spectral matrix of the first spectrum of the test solution into the function model P1 to obtain the model score T1. M and load Q1 M ; (7) Substitute the spectral matrix of the second spectrum of the test liquid into the function model P2 to obtain the model score T2. M and load Q2 M Substitute into the calculation model Y=T1 M ×Q1 M +T2 M ×Q2 M The content Y of poly-α-olefin in the test solution was obtained. The first spectrum is the mid-infrared spectrum, and the second spectrum is the Raman spectrum.

2. The method according to claim 1, wherein, In step (1), the number of standard poly-α-olefin samples is ≥50.

3. The method according to claim 2, wherein, In step (1), the number of standard poly-α-olefin samples is 60-200.

4. The method according to claim 1 or 2, wherein, In step (1), the poly-α-olefin content in the standard poly-α-olefin sample meets the following requirement: mass concentration greater than 1 wt%.

5. The method according to claim 4, wherein, In step (1), the poly-α-olefin content in the standard poly-α-olefin sample meets the following requirement: mass concentration greater than 50-99.8 wt%.

6. The method according to claim 1, wherein, In step (2), the spectral acquisition conditions include: acquisition temperature of 15-50℃; And / or, the acquisition resolution is 2-16 cm. -1 .

7. The method according to claim 6, wherein, In step (2), the spectral acquisition conditions include: acquisition temperature of 23-27℃; And / or, the acquisition resolution is 5-12 cm. -1 .

8. The method according to claim 1, wherein, In step (2), the wavenumber range for mid-infrared spectroscopy is 650-4000 cm⁻¹. -1 .

9. The method according to claim 1, wherein, In step (2), the wavenumber range for Raman spectroscopy acquisition is 100-2500 cm⁻¹. -1 .

10. The method according to claim 1, wherein, Before selecting the characteristic spectral bands, the method further includes preprocessing the spectrum to eliminate redundant information and / or noise in the spectrum.

11. The method according to claim 10, wherein, The preprocessing method is selected from at least one of second-order differential, multivariate scattering correction, first-order differential, and standard normal variable transformation.

12. The method according to claim 11, wherein, The mid-infrared spectrum was preprocessed using second-order differential, and the Raman spectrum was preprocessed using multivariate scattering correction.

13. The method according to claim 1, wherein, The wavenumber range of the characteristic spectral band of the mid-infrared spectrum is 650-4000 cm⁻¹. -1 ; And / or, the wavenumber range of the characteristic spectral bands of Raman spectroscopy is 100–2500 cm⁻¹. -1 .

14. The method according to claim 13, wherein, The wavenumber range of the characteristic spectral band in the mid-infrared spectrum is 1603-1663 cm⁻¹. -1 and 650-980 cm -1 ; And / or, the wavenumber range of the characteristic spectral bands of Raman spectroscopy is 1018–1755 cm⁻¹. -1 .

15. The method according to claim 1 or 10, wherein, In step (4), the orthogonal method is selected from Schmidt orthogonalization and / or normal orthogonalization.

16. The method according to claim 15, wherein, In step (4), the orthogonal method is selected from Schmidt orthogonalization.

17. The method according to claim 1, wherein, The sample was a polyalphaolefin synthetic oil.

Images