A method for identifying olive oil grades by establishing a chemometric model
Through gas-phase ion migration analysis and stoichiometric model, PCA and PLS-DA are used to identify volatile compounds in olive oil, and the objectivity and accuracy of olive oil grade identification is solved, achieving rapid and accurate olive oil grade Identification and high-precision assessment.
Patent Information
- Application Number
- CN202210267119.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-03-18
- Publication Date
- 2024-12-27
- Estimated Expiration
- 2042-03-18
AI Technical Summary
The prior art is difficult to identify the grade of olive oil objectively and accurately from the molecular level, it is easily disturbed by external factors, and sensory analysis is subjective.
The gas-phase ion migration analysis method was used to determine volatile compounds in olive oil, and qualitative analysis was carried out in combination with NIST and IMS databases, and stoichiometric model was established, and the identification was carried out through principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA). Model to realize objective assessment of olive oil grade.
It achieves rapid and accurate identification of olive oil grades, avoids interference from foreign solvents, improves detection accuracy, and has good fit and prediction capabilities.
Smart Images
Figure SMS_1 
Figure SMS_2 
Figure SMS_3
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of analytical chemistry testing and relates to a method for identifying olive oil grades by establishing a chemometrics model. Background Art
[0002] Olive oil is cold-pressed from fresh olive fruits only through mechanical or centrifugal physical methods. It is known as the "Queen of Vegetable Oils, Liquid Gold" in the West for its unique flavor, rich nutrients and functional active ingredients. Olive oil is generally divided into extra virgin olive oil, refined olive oil and pomace oil based on physical, chemical and sensory indicators. In terms of sensory characteristics, olive oil is evaluated and analyzed by a group of specially trained experts, so its analysis process is usually subjective and easily interfered by external factors. It is difficult to study and analyze at the molecular level, so it is of certain research significance to seek objective analytical chemistry methods.
[0003] Extra virgin olive oil usually refers to olive oil with an acidity of no more than 0.8%, without any preservatives or additives, and its physical and chemical indicators and sensory characteristics must also meet the relevant olive oil standards. It is the highest grade and highest quality olive oil, a pure natural product, with the functions of preventing cardiovascular and cerebrovascular diseases, diabetes and other diseases, and improving the function of the digestive system. The free acidity of refined olive oil does not exceed 0.3%. It is an oil product made from virgin olive oil lamp oil through a refining process (its glyceride structure must not be changed). Compared with extra virgin oil, it exhibits a small amount of peculiar smell; pomace oil, an oil obtained from olive pomace using solvents or other physical methods, excluding oils obtained through heavy esterification processes, has sensory indicators that exceed the standard and does not have the good flavor attributes of olive oil.
[0004] Chemometrics includes unsupervised and supervised methods for multivariate data analysis. Principal component analysis (PCA) is an unsupervised pattern recognition technique. This is an exploratory analysis that aims to reduce the dimension and extract the most relevant information. Through the method of data dimensionality reduction, the information of the original data can be retained to the greatest extent. Partial least squares discriminant analysis (PLS-DA) is a supervised pattern recognition technique that can accurately classify and predict samples and screen out differential markers. It is a model method. By constructing a classification prediction model, it can be further used to identify the attribution of more samples, which is not possible with the exploratory PCA method. Summary of the invention
[0005] In view of the above problems, the object of the present invention is to provide a method for identifying extra virgin olive oil, refined olive oil and pomace oil by establishing a chemometric model, thereby objectively analyzing and evaluating the grade of olive oil.
[0006] The present invention adopts a gas phase ion migration analysis method to determine the volatile compounds in known and unknown olive oil samples to be tested. The NIST database and IMS database equipped with the application software can perform qualitative analysis on the signal peak, identify the volatile compounds contained in the olive oil, and quantify the peak volume of the compound. The peak volume information of the volatile compounds contained in the olive oil sample of known grade is used as the analysis data of the chemometric model, imported into the SIMCA14.1 analysis software, and an unsupervised principal component analysis (PCA) and a supervised partial least squares discriminant analysis (PLS-DA) identification model are established. In order to verify the authenticity and reliability of the established model, the PLS-DA model was subjected to 200 random permutation tests. The R2 and Q2 in the original PLS-DA model were higher than the values in the permutation model. The regression line was located at the Q2 point after the permutation and intersected with the vertical axis and was lower than zero, indicating that the PLS-DA model has good fitting and prediction capabilities; the peak volume information of the volatile compounds contained in the olive oil sample of unknown grade is substituted into the above identification model for analysis, and the grade of the olive oil sample can be determined. The specific steps are as follows:
[0007] (1) Pretreatment of olive oil samples of known grade and unknown grade: Place the olive oil samples of known grade and unknown grade in headspace injection bottles, respectively, and incubate them to allow the volatile components in the samples to evaporate into the headspace injection bottles. The olive oil samples of known grade include extra virgin olive oil, refined olive oil, and pomace oil. A total of 29 samples of olive oil samples of known grade were selected, including 25 extra virgin olive oils, 3 refined olive oils, and 1 pomace oil; the incubation speed was 400-600 rpm, the incubation temperature was 70-90°C, and the incubation time was 10-20 min.
[0008] (2) Detection of olive oil samples of known grade and olive oil samples of unknown grade: The olive oil samples of known grade and olive oil samples of unknown grade after the above incubation are separated and detected on a gas phase ion mobility spectrometer to obtain spectra of volatile compounds in all olive oil samples, identify the volatile compounds contained in the olive oil samples, and quantify the peak volumes of volatile compounds in the olive oil samples of known grade and the olive oil samples of unknown grade by using an area normalization method;
[0009] The gas phase ion mobility spectrometry instrument conditions are as follows:
[0010] Chromatographic column: MXT-5 (0.53mmI.D.×15m, df=1.0μm); column temperature: 60℃; carrier gas flow rate: 0-2min, 2mL / min, 2-20min, 2mL / min-100ml / min; carrier gas / drift gas: N2 (99.999%); injection needle temperature: 85℃; injection volume: 500 μL; ion mobility spectrometry conditions: IMS temperature: 45℃; drift gas flow rate: 150mL / min. MXT-5 is a weak polar column, the filler in the column is cross-linked 5% diphenyl and 95% dimethyl polysiloxane, and MXT-5 is a stainless steel capillary column with fused quartz coating. The gas phase ion mobility spectrometer uses the NIST database and IMS database equipped with the software to qualitatively analyze the signal peaks of the spectrum of volatile compounds and identify the volatile compounds contained in the olive oil sample.
[0011] (3) Establishment of grade identification model: The peak volume information of volatile compounds in olive oil samples of known grades was used as chemometric analysis data. The principal component analysis and partial least squares method in SIMCA14.1 analysis software were used to perform dimensionality reduction and cluster analysis on the peak volume data of volatile compounds in olive oil samples of known grades. The grade identification model - unsupervised principal component analysis model and partial least squares discriminant analysis identification model was established.
[0012] (4) Grade identification of olive oil samples of unknown grade: Substitute the peak volume information of volatile compounds contained in the olive oil samples of unknown grade into the grade identification model established in step (3) for analysis to determine the grade of the sample.
[0013] The beneficial effects of the present invention are as follows: the present invention only requires simple headspace incubation pretreatment, avoids interference from external solvents, and increases the accuracy of detection. The method is simple and fast to operate, objectively and accurately evaluates olive oil, and can quickly identify olive oils of different grades. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] Figure 1 is the unsupervised principal component analysis (PCA) discriminant model score plot;
[0015] Figure 2 is the supervised partial least squares discriminant analysis (PLS-DA) discriminant model score plot;
[0016] Figure 3 It is the permutation test plot of the PLS-DA model. DETAILED DESCRIPTION
[0017] The method for identifying the grade of olive oil by establishing a chemometric model of the present invention is further described below through specific implementation. The following examples are used to illustrate the present invention, but are not used to limit the scope of the present invention. Example
[0018] The method for identifying the grade of olive oil by establishing a chemometric model in the present invention specifically comprises the following steps:
[0019] (1) Selecting olive oil samples of known grades for model building: A total of 29 olive oil samples of known grades were selected for model building, as shown in Table 1. Among them, OO-1 to OO-14 and OO-18 to OO-28 are extra virgin olive oils, originating from China; OO-15 to OO-17 are refined olive oils, originating from Spain and Italy; OO-29 is olive pomace oil, originating from China. The specific information of the above olive oil samples is shown in Table 1.
[0020] (2) Pretreatment of olive oil samples of known grade: Take three portions of each olive oil sample in step (1), each portion is 1 mL, and place the olive oil samples to be tested in 20 mL headspace injection bottles. Incubate at an incubation speed of 500 rpm and an incubation temperature of 80°C for 15 min to allow the volatile components in the samples to evaporate as much as possible to the top of the headspace bottle. Incubating the olive oil samples under such incubation conditions will help some components in the olive oil samples to evaporate, and will not destroy the structural information of the olive oil components, and will not affect the accuracy of the results.
[0021] (3) Detection of olive oil samples of known grade: After incubation, the samples were separated and tested on a gas phase ion mobility spectrometer. First, they were separated by a gas chromatography column and then entered into an ion migration tube. After the molecules to be tested were ionized in the ionization zone, they migrated to the Faraday plate for detection under the action of the electric field and the reverse drift gas, achieving secondary separation and obtaining the spectrum of the volatile flavor substances in the olive oil to be tested. The NIST database and IMS database equipped with the application software can perform qualitative analysis on the volatile compounds and identify the volatile compounds contained in the olive oil samples, as shown in Table 2 (each olive oil sample contains the compounds in Table 2). The listed compounds were quantified by area normalization method. The peak volume of each compound in all samples was imported into Excel sheet 1. The first column of data was the name of the olive oil sample, and each subsequent column was the peak volume content of each volatile compound in the olive oil sample (rows 1-75 in the Excel sheet corresponded to the data of extra virgin olive oil (a total of 75 sets of data were measured from 25 samples), rows 76-84 corresponded to the data of refined olive oil (a total of 9 sets of data were measured from 3 samples), and rows 85-87 corresponded to the data of pomace oil (a total of 3 sets of data were measured from 1 sample).
[0022] Instrument conditions: Chromatographic column: MXT-5 (0.53mmI.D.×15m, df=1.0μm); column temperature: 60℃; carrier gas flow rate: 0-2min, 2mL / min, 2-20min, 2mL / min-100ml / min; carrier gas / drift gas: N2 (99.999%); injection needle temperature: 85℃; injection volume: 500 μL; ion mobility spectrometry conditions: IMS temperature: 45℃; drift gas flow rate: 150mL / min;
[0023] MXT-5 is a weak polar column. The filler inside the column is cross-linked 5% diphenyl and 95% dimethylpolysiloxane. It is a stainless steel capillary column with a fused quartz coating.
[0024] (4) Pretreatment of olive oil samples of unknown grade: Take three samples of olive oil of unknown grade (the olive oil samples of unknown grade to be tested are shown in Table 3 Test-1 to Test-5), each with 1 mL, and place them in 20 mL headspace injection bottles. Incubate at an incubation speed of 500 rpm and an incubation temperature of 80 °C for 15 min to allow the volatile components in the sample to evaporate as much as possible into the headspace bottle.
[0025] (5) Detection of unknown grade olive oil samples: After the unknown grade olive oil samples are incubated, they are separated and detected under the conditions of the gas phase ion mobility spectrometer in step (3) to determine the volatile compounds in the unknown grade olive oil, perform qualitative and quantitative analysis, and obtain the corresponding peak volumes. The peak volume information data of the compounds contained in the samples are imported into Excel sheet 2. The first column of the data is the serial number of the unknown grade olive oil samples, and each subsequent column is the peak volume content of each volatile compound in the sample.
[0026] (6) Establishment of grade identification model: The peak volume information of volatile compounds in olive oil samples of known grades was used as the analytical data of the chemometric model. The principal component analysis and partial least squares method in SIMCA14.1 analysis software were used to perform dimensionality reduction and cluster analysis on the above data to establish a grade identification model. The specific operations are as follows:
[0027] Import the peak volume data of volatile compounds in the known grade olive oil samples in Excel sheet 1 and the peak volume data of volatile compounds in the unknown grade olive oil samples in Excel sheet 2 into the SIMCA14.1 software according to the steps of "File—New—regular project—Select data". In the New of the operation interface, select Sheet1 in Select data, and then in New observations, set extra virgin olive oil to class1, refined olive oil to class2, pomace oil to class3, Specify transformation in Transform to Log, type in Scale to UV, and Model type to PCA-X, and obtain an unsupervised principal component analysis (PCA) model, as shown in the figure. Figure 1 , with a contribution rate of 51.6%. Figure 1 It can be seen that the three types of oil are clearly distinguished. In order to increase the classification, a supervised partial least squares discriminant analysis (PLS-DA) identification model is established. The operation steps are the same as those for establishing the principal component model analysis. Only the Model type is set to PLS-DA at the end. In this way, the partial least squares discriminant analysis model diagram is obtained ( Figure 2 ),from Figure 2 To verify the authenticity and reliability of the model, 200 random permutation tests were performed on the PLS-DA model, namely the Permutations analysis interface in the Analyze operation interface. The results are shown in the figure below. Figure 3 As shown, R2 (0.0, 0.0959) and Q2 (0.0, -0.472), which means that the R2 and Q2 in the original PLS-DA model are higher than those in the permuted model, and the regression line intersects the vertical axis at the Q2 point after permutation and is below zero, indicating that the PLS-DA model has good fitting and prediction capabilities.
[0028] (7) Identification of unknown grade olive oil samples: Based on the grade identification model established in step (6), follow the steps “Predict—Specify—Select all unknown grade olive oil samples (Test-1 to Test-5) in available observations and add them to Include observations”. After confirmation, click on classification list to find out which grade each unknown olive oil sample belongs to, and then determine the grade of the olive oil sample. It is determined that the samples Test-1 to Test-5 are all extra virgin olive oil.
[0029]
[0030]
[0031]
[0032]
[0033] From the data in Table 2, it can be seen that the compounds are divided into monomers and dimers, which is related to the proton affinity and concentration of the compounds. The higher the affinity and concentration, the easier it is to form dimers.
[0034]
Claims
1. A method for identifying the grade of olive oil by establishing a chemometric model, comprising the following steps: (1) Pretreatment of olive oil samples of known grade and olive oil samples of unknown grade: placing the olive oil samples of known grade and olive oil samples of unknown grade in headspace injection bottles respectively, and incubating them to allow volatile components in the samples to evaporate into the headspace injection bottles, wherein the olive oil samples of known grade include extra virgin olive oil, refined olive oil and pomace oil; the incubation speed is 400-600 rpm, the incubation temperature is 70-90° C., and the incubation time is 10-20 min; (2) Detection of olive oil samples of known grade and olive oil samples of unknown grade: The olive oil samples of known grade and olive oil samples of unknown grade after the above incubation are separated and detected on a gas phase ion mobility spectrometer to obtain spectra of volatile compounds in all olive oil samples, identify the volatile compounds contained in the olive oil samples, and quantify the peak volumes of volatile compounds contained in the olive oil samples of known grade and olive oil samples of unknown grade; The gas phase ion mobility spectrometry instrument conditions are as follows: Chromatographic column: MXT-5; Column temperature: 60℃; carrier gas flow rate: 0-2min, 2mL / min, 2-20min, 2mL / min-100ml / min; carrier gas / drift gas: N2 (99.999%); injection needle temperature: 85℃; injection volume: 500 μL; ion mobility spectrometry conditions: IMS temperature: 45℃; drift gas flow rate: 150mL / min; The specification of MXT-5 is 0.53mmI.D.×15m, df=1.0μm. MXT-5 is a weak polar column. The filler in the column is cross-linked 5% diphenyl and 95% dimethyl polysiloxane. MXT-5 is a stainless steel capillary column with fused quartz coating. (3) Establishment of grade identification model: The peak volume information of volatile compounds in olive oil samples of known grades was used as chemometric analysis data. The principal component analysis and partial least squares method in SIMCA14.1 analysis software were used to perform dimensionality reduction and cluster analysis on the peak volume data of volatile compounds in olive oil samples of known grades. The grade identification model - unsupervised principal component analysis model and partial least squares discriminant analysis identification model was established. (4) Grade identification of olive oil samples of unknown grade: Substitute the peak volume information of volatile compounds contained in the olive oil samples of unknown grade into the grade identification model established in step (3) for analysis to determine the grade of the sample.
2. The method for identifying the grade of olive oil by establishing a chemometric model according to claim 1, characterized in that: In step (1), a total of 29 samples of olive oil with known grades were selected, including 25 extra virgin olive oils, 3 refined olive oils, and 1 pomace oil.
3. The method for identifying the grade of olive oil by establishing a chemometric model according to claim 1, characterized in that: In step (2), the gas phase ion mobility spectrometer uses the NIST database and IMS database equipped with the software to perform qualitative analysis on the signal peaks of the spectrum of volatile compounds, and identifies the volatile compounds contained in the olive oil sample.