A method for analyzing chlorophyll a and b and beta-carotene in plant extracts

Abstract

The application discloses a method for analyzing chlorophyll a, chlorophyll b and beta-carotene in plant extract, comprising the following steps: S1, using an ultraviolet-visible spectrometer to measure visible spectra of plant sample extract; S2, arranging ultraviolet spectrum data of multiple samples to obtain a bilinear mixed signal number array D, and using MCR-ALS algorithm to decompose the array into a relative concentration change profile matrix C and a normalized spectrum profile matrix S T , and then selecting a pure concentration change profile vector c n and a pure spectrum profile vector s n of the first n single analyte; T S3, for the resolved first n component, using the pure concentration change profile vector c n and the pure spectrum profile vector s n T , reconstructing a pure absorbance matrix Dn=c n ◦s n T ; S4, calculating the concentration: calculating the concentration according to the Lambert-Beer law. The application not only inherits the advantages of the Arnon formula, but also can overcome spectral interference to give more accurate quantitative results.

Description

Technical Field

[0001] This invention relates to a method for analyzing chlorophyll a and b and β-carotene in plant extracts, belonging to the technical field of plant chlorophyll and carotene analysis. Background Technology

[0002] Life on Earth ultimately depends on energy from the sun, and photosynthesis, the only biological pathway for plants, algae, and photosynthetic bacteria to obtain this energy, is through photosynthesis. The mesophyll cells of plant leaves contain numerous chloroplasts, the site of photosynthesis. Their inner membrane branching system contains photosynthetic pigments that absorb light energy. These pigments are primarily chlorophyll and bacterial chlorophyll, but also include various carotenoids (such as β-carotene) and bile pigments (such as phycoerythrin and phycocyanin). Green plants contain large amounts of chlorophyll a and b, while prokaryotes and cyanobacteria contain chlorophyll c and d. Carotenoids act as co-pigments, transferring absorbed light energy to chlorophyll for photosynthesis and helping to protect against light damage. Chlorophyll converts light energy into chemical energy through two different photosystems. In plant photosynthesis, solar energy is absorbed by the plant's photosynthetic pigments, oxidizing water to produce oxygen and reducing carbon dioxide to synthesize carbohydrates (mainly sugars). The energy stored in these carbohydrate molecules can then power plant cells and serves as an energy source for all living organisms. Therefore, it is of great significance to determine the content of chlorophyll a and b and β-carotene in plant systems (e.g., leaves), and it is also frequently required in agricultural and forestry sciences.

[0003] For determining the content of chlorophyll a, b, and β-carotene in complex plant extracts, high-performance liquid chromatography (HPLC) and even liquid chromatography-mass spectrometry (LC-MS) are excellent separation and analytical strategies. If accurate quantitative analysis of chlorophyll a, b, and β-carotene in plant systems is required in the laboratory, HPLC and LC-MS are powerful and highly accurate analytical strategies. However, if rapid on-site detection is needed (e.g., at field trials or baking sites), HPLC and LC-MS are limited by factors such as high instrument costs, demanding maintenance requirements, and the need for expensive analytical standards.

[0004] Studies have shown that chlorophyll a and b, as well as β-carotene, all possess large conjugated systems. The maximum absorption wavelengths of chlorophyll a and b in the red light region are located at 663 nm and 645 nm, respectively, while the absorption range of β-carotene is 400-500 nm. They have large absorption coefficients, making UV-Vis spectrophotometry a good method for measuring them. Daniel I. Arnon proposed the famous Arnon formula for determining chlorophyll a and b in 1949. This formula is based on the Lambert-Beer law at wavelengths of 663 and 645 nm, using the absorption coefficients measured by G. MacKinney in 1941 to directly calculate the concentrations of chlorophyll a and b. Subtracting the absorbance of chlorophyll a and b at a wavelength of 470 nm yields the formula for calculating the concentration of β-carotene according to the Lambert-Beer law. The advantages of the Arnon formula include: the UV-Vis spectrophotometer is inexpensive, requires minimal maintenance, and does not require expensive analytical standards. Therefore, it is ideally suited for rapid on-site analysis tasks in agricultural and forestry science and engineering, and remains the most commonly used and classic method for determining chlorophyll a and b, as well as β-carotene, in plant extracts. In 2021, Weksu et al. proposed a rapid, non-destructive method for determining chlorophyll and β-carotene using a fiber optic near-infrared spectrometer; this method still requires basic concentration data based on the Arnon formula. Summary of the Invention

[0005] Research has shown that the Arnon formula faces the challenge of spectral interference, which introduces varying degrees of error, leading to inaccurate measurement results. Therefore, this invention provides a method for analyzing chlorophyll a and b, as well as β-carotene, in plant extracts, using tobacco leaf extract as an example. This method aims to solve the spectral interference problem encountered when determining chlorophyll a, b, and β-carotene in plant systems using ultraviolet-visible spectrophotometry.

[0006] The technical solution of this invention is: a method for analyzing chlorophyll a and b and β-carotene in plant extracts, comprising:

[0007] S1, Measure the visible spectrum: Use an ultraviolet-visible spectrometer to measure the visible spectrum of extracts from multiple plant samples;

[0008] S2, Analyte spectra are resolved using MCR-ALS: The UV spectral data of multiple samples are arranged row by row to obtain a bilinear mixed signal matrix D. The MCR-ALS algorithm is used to decompose it into a relative concentration change profile matrix C and a normalized spectral profile matrix S. T Then select the pure concentration change profile vector c of the nth analyte. n and pure spectral profile vector s n T ;

[0009] S3, Reconstruct the pure absorbance matrix of a single analyte: For the resolved nth component, use the pure concentration change profile vector c that is resolved for it. n and pure spectral profile vector s n T The pure absorbance matrix is ​​reconstructed through the vector outer product operation. That is, the pure spectral matrix X of chlorophyll a, the pure spectral matrix Y of chlorophyll b, and the pure spectral matrix Z of β-carotene are reconstructed;

[0010] S4, Calculate the concentration: Select the quantitative wavelengths corresponding to chlorophyll a and b and β-carotene and refer to or measure the absorbance coefficients at those wavelengths. Then, calculate the concentrations according to the Lambert-Beer Law A = Kb c, where A is the reconstructed pure absorbance, b is the thickness of the absorption layer, c is the concentration of the absorbing substance, and K is the absorbance coefficient.

[0011] Optionally, the construction process of the bilinear mixed signal array D is as follows:

[0012] If a UV-Vis spectrometer is used to measure the absorbance of a sample, containing J wavelength points, then the spectral data of I samples, arranged row by row, can construct a number matrix D containing I samples × J variables; according to the Lambert-Beer law, each element d in this number matrix D... ij It can be represented as follows:

[0013]

[0014] Among them, c in s represents the concentration of the nth substance in the i-th sample. jn c represents the absorption coefficient of the nth substance at the jth wavelength. in and s jn They are number arrays D I×J The relative concentration change profile matrix C I×N and spectral profile matrix S J×N The elements in the i-th row and n-th column and the j-th row and n-th column; N represents the number of components with light absorption; d ij For c respectively in and s jn All of them have a linear relationship, therefore this sequence conforms to the bilinear model; in scalar terms, this bilinear matrix can be represented as follows:

[0015]

[0016] Based on the terminology of vector outer product, it can be expressed as follows:

[0017]

[0018] Wherein, column vector c .nThe row vector s represents the concentration change profile of the nth substance in the sample dimension. .n T This represents the absorption coefficient profile of the nth substance in the variable dimension, i.e., the absorption spectrum;

[0019] From the perspective of the contour matrix, it can be represented as follows:

[0020]

[0021] That is, D I×J =C I×N S J×N T

[0022] Optionally, the MCR-ALS algorithm can be used to decompose matrix D into a relative concentration change profile matrix C and a normalized spectral profile matrix S. T The process is as follows:

[0023]

[0024] Among them, C and S T These are matrices composed of the contours of N components in the row and column directions, respectively, and E is the residual matrix.

[0025] Optionally, the chlorophyll a and b, and β-carotene are calculated using the following formulas:

[0026] c i,Chla =X i,663 / (K Chla,663 ·b)(g L -1 )

[0027] c i,Chlb =Y i,645 / (K Chlb,645 ·b)(g L -1 )

[0028] c i,β-Carotene =Z i,452 / (K β-Carotene,452 ·b)(g L -1 )

[0029] In the formula, X, Y, and Z are the reconstructed pure spectral matrices of chlorophyll a, chlorophyll b, and β-carotene, respectively. i,663 Y i,645 Z i,452 Let c represent the pure absorbance of chlorophyll a, chlorophyll b, and β-carotene in the i-th sample at 663 nm, 645 nm, and 452 nm, respectively. i,Chla c i,Chlb c i,β-CaroteneLet represent the concentrations of chlorophyll a, chlorophyll b, and β-carotene in the i-th sample, respectively.

[0030] The beneficial effects of this invention are as follows: By introducing multivariate curve resolution-alternating least squares (MCR-ALS) algorithm to assist visible spectroscopy, this invention can distinguish and extract the pure signals of chlorophyll a and b from the mixed signals of the analytical system, thus overcoming the problem of incompletely selective quantification. Then, based on Lambert-Beer's law, the concentration is calculated, and a rapid, simple, low-cost, interference-resistant, and accurate quantitative analytical method for determining chlorophyll a, b, and β-carotene in plant extracts is established. This method not only inherits the advantages of Arnon's formula, but also overcomes potential spectral interference in the analytical system, and is expected to provide a highly selective determination method for the quantitative analysis of chlorophyll a, b, and β-carotene in plant extracts. Attached Figure Description

[0031] Figure 1 UV-Vis spectra of chlorophyll a and b and β-carotene and Arnon's formula;

[0032] Figure 2 Spectral interferences that may exist when determining chlorophyll a, b, and β-carotene based on the Arnon formula;

[0033] Figure 3 Schematic diagram of a method for accurate determination of chlorophyll a, b and β-carotene using MCR-ALS-assisted visible spectroscopy with anti-interference capability;

[0034] Figure 4 Box plots of chlorophyll a, b, and β-carotene concentration changes in leaves (upper tobacco leaves from 2022, variety Yunyan 87, origin Qianxi City) during the curing process, predicted by visible spectrum and MCR-ALS. The box plots, from top to bottom, represent the upper edge, upper quartile, median, lower quartile, and lower edge, with outliers indicated by "+". Subplots (a), (b), and (c) on the left show the concentration changes of chlorophyll a, b, and β-carotene during curing with spring-loaded tobacco sticks; subplots (d), (e), and (f) on the right show the concentration changes of chlorophyll a, b, and β-carotene during curing with tobacco clips. Detailed Implementation

[0035] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0036] Example 1: Discovery of the Arnon Formula Assumption

[0037] Lambert-Beer Law: When a beam of monochromatic light passes through a solution containing an absorbing substance, the absorbance of each absorbing substance is directly proportional to the concentration of that substance and the thickness of the absorbing layer. This law can be expressed by the formula:

[0038] A = K bc (1)

[0039] Where A is absorbance, b is the thickness of the absorption layer, c is the concentration of the absorbing substance, and K is the absorption coefficient. When b is in cm and c is in g / L... -1 When K is in L g -1 cm -3 In the above law, absorbance exhibits a typical single linear characteristic with respect to concentration; the absorbance of the solution is equal to the sum of the absorbances of all absorbing substances in the solution.

[0040] The maximum absorption wavelengths of chlorophyll a and b in the red light region are 663 nm and 645 nm, respectively. If only chlorophyll a and b absorb incident light at these two wavelengths, the following system of two linear equations can be established according to the Lambert-Beer law:

[0041] A 663 =K Chla,663 bc Chla +K Chlb,663 bc Chlb (2)

[0042] A 645 =K Chla,645 bc Chla +K Chlb,645 bc Chlb (3)

[0043] Among them, A 663 and A 645 The absorbance of the solution at 663 nm and 645 nm are respectively, c Chla and c Chlb These represent the concentrations (g / L) of chlorophyll a (Chla) and chlorophyll b (Chlb), respectively. -1 The absorption layer thickness b is 1 cm. The absorption coefficients K of chlorophyll a and b at 663 nm are... Chla,663 and K Chlb,663 The values ​​were 82.04 and 9.27 L g, respectively. -1 cm -3 The absorbance coefficients K of chlorophyll a and b at 645 nm Chla,645 and K Chlb,645 The values ​​were 16.75 and 45.60 L g, respectively. -1 cm-3 .

[0044] Substituting the above values ​​into formulas (2) and (3), we can obtain the following system of two linear equations in two variables (using only numerical calculations):

[0045] A 663 =82.04 c Chla +9.27 c Chlb (4)

[0046] A 645 =16.75 c Chla +45.60 c Chlb (5)

[0047] Solving the above system of two linear equations using the elimination method yields the calculated concentrations of chlorophyll a and b (g / L). -1 Arnon's formula:

[0048] c Chla =(12.72A) 663 -2.59A 645 ) / 1000 (6)

[0049] c Chlb =(22.88 A) 645 -4.67A 663 ) / 1000 (7)

[0050] If only chlorophyll a and b, and β-carotene absorb incident light at 470 nm, then the concentration of β-carotene (g / L) is... -1 It can be calculated using the following expression (using only numerical operations):

[0051] c β-Carotene =(A 470 -2.0 c Chla -114.8 c Chlb ) / 245 (8)

[0052] Chlorophyll a and b both have absorption peaks in the blue and red light regions. In the blue light region, the absorption peaks of chlorophyll a and b are located in the 400-500 nm range; however, some substances in plants (such as carotenoids, flavonoids, and cytochromes) have absorption signals in this region, resulting in significant spectral interference. In the red light region, the maximum absorption wavelengths of chlorophyll a and b are at 663 nm and 645 nm, respectively, where there are fewer interfering substances and the interference is less severe. Therefore, Daniel I. Arnon chose wavelengths of 663 nm and 645 nm and, based on Beer-Lambert's law, established a linear equation in two variables concerning the concentrations of chlorophyll a and b, as follows: Figure 1As shown. The Arnon formula for directly calculating the concentrations of chlorophyll a and b based on their absorption coefficients can be obtained using the elimination method. This method is inexpensive, easy to operate, and simple to use. Obviously, the Arnon formula for calculating the concentrations of chlorophyll a and b has an important assumption: that only chlorophyll a and b absorb incident light at wavelengths of 663 and 645 nm. Only in this way can a linear equation containing only the concentrations of chlorophyll a and b be derived based on Lambert-Beer's law.

[0053] The Arnon formula selects the red light region for measuring chlorophyll a and b, thus avoiding the absorption peaks of many substances in the plant system. However, some substances still have a certain absorbance in this region, such as bacterial chlorophyll a (this interfering substance may be present if the plant system contains bacteria), chlorophyll d, phycocyanin, etc. Figure 2 As shown, when measuring β-carotene concentration at a wavelength of 470 nm, more and more severe interferences are encountered, such as chlorophyll d, phycoerythrin, xanthophyll, cytochrome c, and flavonoids.

[0054] The magnitude of the error introduced by spectral interference depends on three factors: (1) the quantity of the interfering substance; (2) the concentration of each interfering substance; and (3) the absorption coefficient of each interfering substance at the aforementioned wavelengths. Therefore, attention should be paid to the problem of spectral interference when using the Arnon formula. However, the difficulty lies in the fact that measuring the absorbance at two wavelengths is almost insufficient to determine whether spectral interference exists and its degree.

[0055] Researchers developed a similar formula for calculating β-carotene concentration based on Arnon's formula: Based on the calculated concentrations of chlorophyll a and b, the absorbance of chlorophyll a and b at a wavelength of 470 nm is calculated using the absorbance coefficient. Then, the absorbance of β-carotene at this wavelength is obtained by subtracting the absorbance of chlorophyll a and b at 470 nm from the absorbance measured at 470 nm. The β-carotene concentration is then calculated according to Lambert-Beer's law. It can be seen that the above method for calculating β-carotene concentration makes two important assumptions: (1) it assumes that the concentrations of chlorophyll a and b calculated based on Arnon's formula are accurate; (2) it assumes that only chlorophyll a, b, and β-carotene absorb incident light at a wavelength of 470 nm.

[0056] Based on the above, it can be seen that the Arnon formula in existing measurement methods makes an important assumption: only chlorophyll a and b absorb incident light at wavelengths of 663 and 645 nm. However, there are still some other light-absorbing substances in plant extracts that absorb light at wavelengths of 663 and 645 nm, such as... Figure 2 As shown, these substances can cause spectral interference, thereby reducing the accuracy of measurements.

[0057] Example 2: Method for analyzing chlorophyll a and b and β-carotene in plant extracts

[0058] The method includes the following steps:

[0059] S1, Measure the visible spectrum: Use an ultraviolet-visible spectrometer to measure the visible spectrum of extracts from multiple plant samples;

[0060] Measurements were performed using an Evolution-201 UV-Vis absorption spectrometer (Thermo Fisher Scientific, Madison, USA), employing 1.00 cm quartz cuvettes and operating at room temperature (25°C). The spectral measurement range was set to 400-700 nm (sampling intervals of 1 nm). The spectral bandwidth was set to 1 nm, the integration time to 0.03 s, and the scan rate to 1200 nm min. -1 .

[0061] Measurements were taken using cured tobacco leaves. Yunyan 87, the main tobacco variety cultivated in Qianxi City, Guizhou Province, was selected as the research subject. Curing and sampling were carried out sequentially on the upper and middle leaves. Before curing, the leaves were initially screened, selecting approximately 200 intact, disease-free leaves of relatively uniform size, shape, and maturity. The selected leaves were tied to poles and marked (two poles were prepared for spring-loaded tobacco and two for tobacco clips, approximately 50 leaves per pole). The marked leaves from both types of tobacco poles were placed in two curing barns (the lower level, on the left and right sides near the door), arranged from the second pole near the door inwards. The optimal curing time was determined to begin.

[0062] During the baking process, samples were taken from both batches of smoke, with the following sampling arrangements: samples were taken sequentially at the following times: ignition, 36℃ (early yellowing stage), 38℃ (early yellowing stage), 40℃ (mid-yellowing stage), 42℃ (late yellowing stage), 44-46℃ (early color fixing stage), 48-49℃ (mid-color fixing stage), 52-54℃ (late color fixing stage), 60℃ (early drying stage), and 68℃ (late drying stage). Five leaves were taken each time.

[0063] For each leaf sample, six sampling zones were taken from the upper, middle, and lower parts on both sides of the main vein using a 6mm diameter perforator. The six perforated samples were placed together in a 5mL centrifuge tube and extracted with 4mL of 95% ethanol solution for 24 hours. Before measuring the visible spectrum, the centrifuge tube was shaken three times to ensure thorough mixing. Using 95% ethanol solution as a blank, the visible spectrum was measured using a UV-Vis spectrometer.

[0064] S2, Analyte spectra are resolved using MCR-ALS: The UV spectral data of multiple samples are arranged row by row to obtain a bilinear mixed signal matrix D. The MCR-ALS algorithm is used to decompose it into a relative concentration change profile matrix C and a normalized spectral profile matrix S. TThen select the pure concentration change profile vector c of the nth analyte. n and pure spectral profile vector s n T ;

[0065] After obtaining the visible spectra of multiple samples, a bilinear mixed signal array D is first constructed, as follows:

[0066] If a UV-Vis spectrometer is used to measure the absorbance of a sample, containing J wavelength points, then the spectral data of I samples, arranged row by row, can construct a number matrix D containing I samples × J variables; according to the Lambert-Beer law, each element d in this number matrix D... ij It can be represented as follows:

[0067]

[0068] Among them, c in s represents the concentration of the nth substance in the i-th sample. jn c represents the absorption coefficient of the nth substance at the jth wavelength. in and s jn They are number arrays D I×J The relative concentration change profile matrix C I×N and spectral profile matrix S J×N The elements in the i-th row and n-th column and the j-th row and n-th column; N represents the number of components with light absorption; d ij For c respectively in and s jn All of them have a linear relationship, therefore this sequence conforms to the bilinear model; in scalar terms, this bilinear matrix can be represented as follows:

[0069]

[0070] Based on the terminology of vector outer product, it can be expressed as follows:

[0071]

[0072] Wherein, column vector c .n The row vector s represents the concentration change profile of the nth substance in the sample dimension. .n T This represents the absorption coefficient profile of the nth substance in the variable dimension, i.e., the absorption spectrum;

[0073] From the perspective of the contour matrix, it can be represented as follows:

[0074]

[0075] That is, D I×J =C I×N SJ×N T

[0076] Secondly, the MCR-ALS algorithm is used to decompose matrix D into a relative concentration change profile matrix C and a spectral profile matrix S. T The process is as follows:

[0077]

[0078] Among them, C and S T These are matrices composed of the contours of N components in the row and column directions, respectively, and E is the residual matrix.

[0079] Multivariate Curve Resolution-Alternating Least Squares (MCR-ALS) is one of the most commonly used multivariate curve resolution algorithms. It resolves curves through alternating iterations, and appropriate constraints (such as nonnegativity constraints, unimodal constraints, local rank constraints, closure constraints, hard model constraints, etc.) can be selected as needed to obtain better resolution results. MCR-ALS is well-suited for spectroscopic resolution tasks, and the results it provides have physical meaning and are easy to interpret. MCR-ALS has been used in many instrumental analysis fields. It can be used to decompose the response matrix D of an analytical instrument (a mixed signal of all responding components in the sample) into signals of individual pure components (e.g., concentration change profiles and spectral profiles), as shown in the equation above.

[0080] The MCR-ALS algorithm mainly consists of the following steps:

[0081] (1) Determine the number of components N in the model. Principal component analysis is a commonly used method for determining the number of components. Generally, different component numbers should be tried to differentiate them;

[0082] (2) Initialization (e.g., selecting an initial estimate for the contour matrix C);

[0083] (3) Based on the estimated value of the contour matrix C, select appropriate constraints and calculate the contour matrix. Among them, (C) + D represents the generalized inverse of matrix C. PCA It is a number matrix reconstructed based on principal component analysis (PCA);

[0084] (4) Based on the contour matrix S T Based on the estimated value, select appropriate constraints and calculate the contour matrix.

[0085] (5) Based on the estimated matrices C and S from the previous iteration loop T Calculate the estimated value of the measurement array D;

[0086] (6) Repeat steps (3)-(5) until the preset convergence criterion (e.g., the relative change of the sum of squared residuals of the logarithmic matrix D fitting is below a certain threshold) or the maximum number of iterations is reached.

[0087] The pure concentration change profile vectors and pure visible spectra of chlorophyll a and b, as well as β-carotene, were obtained by MCR-ALS resolving bilinear models.

[0088] It should be noted that the traditional MCR-ALS algorithm designs and measures standard samples of the pure analyte (true concentration c). ref Given), then the pure concentration change profile vector c of the distinguished standard sample is obtained. MCR cal For the true concentration c ref Perform univariate regression (construct a standard curve) to identify the profile vector c of the pure concentration change of the unknown samples. MCR unk Substituting these values ​​into the regression equation (standard curve) to predict the true concentration of unknown samples is certainly more accurate, but analytical standards for chlorophyll a and b are expensive. For example, the price of chlorophyll a (≥85.0%, chromatographic grade) sold by the domestic reagent supplier Aladdin is approximately 5457 yuan / 5 mg, with a purity of only ≥85.0%; the price of chlorophyll a and b (analytical standard grade) sold by the international reagent supplier Merck is approximately 4302 yuan / 1 mg. Constructing a standard curve generally requires at least 5 mg of analytical standard, so the cost of constructing a standard curve for chlorophyll a and b is approximately 40,000 yuan. From an experimental cost perspective, one reason why the Arnon formula is widely accepted is that it does not require analytical standards when determining chlorophyll a and b. Therefore, the strategy proposed in this method also bypasses the operation of constructing a standard curve based on analytical standards for chlorophyll a and b, inheriting the advantages of the Arnon formula and having the advantage of anti-interference.

[0089] S3, Reconstructing the pure absorbance matrix of a single analyte: For the resolved nth component, the pure absorbance matrix is ​​reconstructed using its resolved pure concentration change and pure spectral profile vector through a vector outer product operation. That is, the pure spectral matrix X of chlorophyll a, the pure spectral matrix Y of chlorophyll b, and the pure spectral matrix Z of β-carotene are reconstructed; in this reconstructed pure absorbance matrix, each row is the pure spectrum of the substance, without containing the signals of coexisting interference in the system under study.

[0090] S4, Calculate the concentration: Select the quantitative wavelengths corresponding to chlorophyll a and b and β-carotene and refer to or measure the absorbance coefficients at those locations. Then, directly calculate the concentrations according to the Lambert-Beer Law A = Kb c, where A is the reconstructed pure absorbance, b is the thickness of the absorption layer, c is the concentration of the absorbing substance, and K is the absorbance coefficient.

[0091] Specifically, chlorophyll a and b, as well as β-carotene, are calculated using the following formulas:

[0092] c i,Chla =X i,663 / (K Chla,663 ·b)(g L -1 )

[0093] c i,Chlb =Y i,645 / (K Chlb,645 ·b)(g L -1 )

[0094] c i,β-Carotene =Z i,452 / (K β-Carotene,452 ·b)(g L -1 )

[0095] In the formula, X, Y, and Z are the reconstructed pure spectral matrices of chlorophyll a, chlorophyll b, and β-carotene, respectively. i,663 Y i,645 Z i,452 Let c represent the pure absorbance of chlorophyll a, chlorophyll b, and β-carotene in the i-th sample at 663 nm, 645 nm, and 452 nm, respectively. i,Chla c i,Chlb c i,β-Carotene Let represent the concentrations of chlorophyll a, chlorophyll b, and β-carotene in the i-th sample, respectively.

[0096] Analysis showed that for samples collected using a spring-loaded tobacco curing device, the similarity of the spectra of chlorophyll a and b, and β-carotene to their standard spectra was 0.9951, 0.9930, and 0.9972, respectively. Qualitatively, MCR-ALS correctly resolved the spectra of these three analytes. This multivariate curve resolution process decomposes the bilinear matrix D into a concentration change profile matrix C and a spectral matrix S (D = CS). T As long as the spectral matrix S is correctly resolved, the resulting concentration change profile matrix C, from a matrix calculation perspective, is certainly accurate and unique. The detection limits for chlorophyll a and b, and β-carotene, are 0.03, 0.02, and 0.15 μg / mL, respectively. -2For samples collected by roasting tobacco using tobacco clips, the similarity of the resolved spectra to their standard spectra for chlorophyll a and b, and β-carotene, were 0.9983, 0.9932, and 0.9916, respectively (all similarity values ​​were above 0.99), and the limits of detection were 0.05, 0.03, and 0.14 μg / mL, respectively. -2 .

[0097] The predicted concentration change trend is shown in Figure 4 The experimental results of using a spring-loaded tobacco curing rod are shown in the three subfigures on the left. The chlorophyll a content in fresh leaves ranged from 2.5 to 4.5 μg / cm³. -3 Approximately (corresponding to a chlorophyll a concentration of 1.0-1.8 μg / mL in the extract) -2 As the baking process progresses, the chlorophyll a content rapidly decreases, reaching almost zero after about 60 hours (two and a half days, during the later stages of yellowing), at approximately 0.05-0.1 μg / cm³. -3 Approximately (corresponding to a concentration of 0.02-0.05 μg / mL in the extract) -1 (Note that this is roughly the lower limit of quantitation for UV-Vis spectroscopy). After this point, until the baking process is complete, the chlorophyll a content remains below the current detection limit for UV-Vis spectroscopy (it can be considered essentially zero). The chlorophyll b content in fresh leaves is approximately 1.4 μg / cm³. -2 Around 108 hours (four and a half days, mid-stage of color fixation), the chlorophyll b content also decreased (but the degradation rate was slower than that of chlorophyll a), eventually dropping to approximately 0.1 μg / cm³. -3 The values ​​below are essentially below the detection limit of UV-Vis spectroscopy and can be considered practically zero. The β-carotene content in fresh leaves is between 0.25 and 0.5 μg / mL. -1 As the baking process progressed, the β-carotene content first increased and then decreased from 0 to 20 hours (early yellowing) and then to 60 hours (late yellowing). It remained essentially unchanged from 60 hours (late yellowing) to 108 hours (mid-color fixation), and then slightly increased again from 108 hours until the end of baking. Throughout the entire baking process, the β-carotene content in the leaves remained generally between 0.2 and 1.5 μg / mL. -1 The range (corresponding to a carotenoid concentration of 0.1-0.6 μg / mL in the extract) -1 The values ​​(within the specified range) are generally above the lower limit of quantitation for UV-Vis spectroscopy, allowing for accurate determination and monitoring. The experimental results of tobacco roasting using tobacco clips are shown in the three sub-figures on the right. The trends of the three analytes show no significant difference from those in the spring-loaded tobacco roasting experiment.

[0098] The study also designed three levels (five replicates each) of spiked validation sets for chlorophyll a, b, and β-carotene, with chlorophyll a spiked at levels of 0.5, 1.0, and 1.5 μg mL. -1 The spiking levels of chlorophyll b were 0.2, 0.4, and 0.6 μg / mL, respectively. -1 The spiking levels of β-carotene were 0.2, 0.4, and 0.6 μg / mL, respectively. -1 The spiked samples were further analyzed using the methods described above. In the experiment of curing tobacco using a spring-loaded tobacco stick, the average relative errors of this method in predicting chlorophyll a and b, and β-carotene in the spiked validation set were 1.9%, 0.2%, and 1.2%, respectively. In the experiment of curing tobacco using a tobacco clip, the average relative errors of this method in predicting chlorophyll a and b, and β-carotene in the spiked validation set were 0.8%, 1.5%, and 3.6%, respectively. These results indicate that the accuracy of this method is satisfactory.

[0099] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for analyzing chlorophyll a and b and β-carotene in plant extracts, characterized in that, include: S1, Measure the visible spectrum: Use an ultraviolet-visible spectrometer to measure the visible spectrum of extracts from multiple plant samples; S2, Analyte spectra are resolved using MCR-ALS: The UV spectral data of multiple samples are arranged row by row to obtain a bilinear mixed signal matrix D. The MCR-ALS algorithm is used to decompose it into a relative concentration change profile matrix C and a normalized spectral profile matrix S. T Then select the pure concentration change profile vector c of the nth analyte. n and pure spectral profile vector s n T ; S3, Reconstruct the pure absorbance matrix of a single analyte: For the resolved nth component, use the pure concentration change profile vector c that is resolved for it. n and pure spectral profile vector s n T The pure absorbance matrix is ​​reconstructed through the vector outer product operation. That is, the pure spectral matrix X of chlorophyll a, the pure spectral matrix Y of chlorophyll b, and the pure spectral matrix Z of β-carotene are reconstructed; S4, Calculate the concentration: Select the quantitative wavelengths corresponding to chlorophyll a and b and β-carotene and refer to or measure the absorbance coefficients at those locations. Then, calculate the concentrations according to the Lambert-Beer Law A = Kbc, where A is the reconstructed pure absorbance, b is the thickness of the absorption layer, c is the concentration of the absorbing substance, and K is the absorbance coefficient.

2. The method for analyzing chlorophyll a and b and β-carotene in plant extracts according to claim 1, characterized in that, The construction process of the bilinear mixed signal array D is as follows: If a UV-Vis spectrometer is used to measure the absorbance of a sample, containing J wavelength points, then the spectral data of I samples, arranged row by row, can construct a number matrix D containing I samples × J variables; according to the Lambert-Beer law, each element d in this number matrix D... ij It can be represented as follows: Among them, c in s represents the concentration of the nth substance in the i-th sample. jn c represents the absorption coefficient of the nth substance at the jth wavelength. in and s jn They are number arrays D I×J The relative concentration change profile matrix C I×N and spectral profile matrix S J×N The elements in the i-th row and n-th column and the j-th row and n-th column; N represents the number of components with light absorption; d ij For c respectively in and s jn All of them have a linear relationship, therefore this sequence conforms to the bilinear model; in scalar terms, this bilinear matrix can be represented as follows: Based on the terminology of vector outer product, it can be expressed as follows: Wherein, column vector c .n The row vector s represents the concentration change profile of the nth substance in the sample dimension. .n T This represents the absorption coefficient profile of the nth substance in the variable dimension, i.e., the absorption spectrum; From the perspective of the contour matrix, it can be represented as follows: That is, D I×J =C I×N S J×N T .

3. The method for analyzing chlorophyll a and b and β-carotene in plant extracts according to claim 1, characterized in that, The MCR-ALS algorithm is used to decompose matrix D into a relative concentration change profile matrix C and a normalized spectral profile matrix S. T The process is as follows: Among them, C and S T These are matrices composed of the contours of N components in the row and column directions, respectively, and E is the residual matrix.

4. The method for analyzing chlorophyll a and b and β-carotene in plant extracts according to claim 1, characterized in that, The chlorophyll a and b, and β-carotene are calculated according to the following formulas: c i,Chla =X i,663 / (K Chla,663 ·b)(g L -1 ) c i,Chlb =Y i,645 / (K Chlb,645 ·b)(g L -1 ) c i,β-Carotene =Z i,452 / (K β-Carotene,452 ·b)(g L -1 ) In the formula, X, Y, and Z are the reconstructed pure spectral matrices of chlorophyll a, chlorophyll b, and β-carotene, respectively. i,663 Y i,645 Z i,452 Let c represent the pure absorbance of chlorophyll a, chlorophyll b, and β-carotene in the i-th sample at 663 nm, 645 nm, and 452 nm, respectively. i,Chla c i,Chlb c i,β-Carotene Let represent the concentrations of chlorophyll a, chlorophyll b, and β-carotene in the i-th sample, respectively.

Images