Method for classifying a product into a plurality of classes using comprehensive two-dimensional gas chromatography

Abstract

This disclosure relates to methods for classifying a product into multiple classes. Specifically, this disclosure relates to methods for classifying a product into multiple classes using comprehensive two-dimensional gas chromatography. This method can be used to classify lubricating oils. Abstract Figure: Figure 4

Description

Title of the invention: Method for classifying a product into a plurality of classes using comprehensive two-dimensional gas chromatography technical field

[0001] This disclosure relates to methods for classifying a product into a plurality of classes. In particular, this disclosure relates to methods for classifying a product into a plurality of classes using comprehensive two-dimensional gas chromatography. This method can be used to classify lubricating oils; in particular, it allows differentiation between recycled and conventional lubricating oils. Prior art

[0002] Given the very high degree of similarity in chemical composition between certain products within a product line, it is quite difficult to differentiate them. For example, in the field of lubricating oils, it is complex to distinguish between recycled base oils (RBOs) and non-recycled oils, also called conventional base oils (CBOs), or original oils. Indeed, recycled oils undergo various re-refining stages, or different chemical treatments to remove various impurities (additives, oxidation products), which makes their composition very similar to that of a conventional oil.

[0003] It is nevertheless necessary to be able to distinguish these products from one another. In the field of lubricating oils, it is generally preferable to add an additive to recycled oil to improve its performance, and it is therefore useful to know whether the oil is recycled or not in order to best adapt the formulation of lubricating oils. Furthermore, from an ecological and reuse perspective, it is also important to know whether the oil is recycled or not. It is also useful to be able to distinguish between lubricating oils formulated from recycled oils and those formulated from conventional oils.

[0004] It is known to analyze products, such as lubricating oils, by gas chromatography (GC) to determine the major compounds in these products. However, this determination of major compounds often does not allow differentiation between products and does not allow determination of whether the oil is recycled or not.

[0005] It would therefore be useful to develop a method for differentiating i) a recycled oil (RBO) from a conventional oil (CBO) and ii) a lubricating oil formulated from recycled oils (i.e. an additively enhanced RBO) of a lubricating oil formulated from conventional oils (i.e. an additively enhanced CBO).

[0006] It would also be useful to develop a method for classifying a lubricating oil in order to determine whether it is a recycled oil or a conventional oil. Summary

[0007] The present disclosure proposes a process using comprehensive two-dimensional gas chromatography (GCxGC) to differentiate i) a recycled oil (RBO) from a conventional oil (CBO) and ii) a lubricating oil formulated from recycled oils (i.e. an additized RBO) from a lubricating oil formulated from conventional oils (i.e. an additized CBO).

[0008] The present disclosure thus relates to a method of classifying a product among a plurality of classes, via a chromatography device (12), the chromatography device (12) comprising a comprehensive two-dimensional gas chromatography module (13) comprising a first column A and a second column B, and capable of separating different compounds of the product according to their volatility and polarity, the chromatography device (12) further comprising a flame ionization detector (14) capable of measuring an intensity of ionization electric current generated for each compound included in the product, the chromatography device (12) being calibrated with at least one calibration product, allowing the retention time of the different compounds present in the product to be corrected; the process being implemented by an electronic classification device (20) and comprising the following steps: a. determine (100) a table describing the intensity of ionization electric current generated for each compound included in the product as a function of the corrected retention times in columns A and B, from a measurement made by the chromatography device (12) on the product; b. assign a class to the product (120), from among a plurality of classes, by applying a multivariate statistical algorithm to the table, said algorithm being trained on tables obtained from reference products.

[0009] Given the complexity of the oil mixtures present in lubricating oils in particular, and the diversity of the chemical natures of the compounds constituting these oils, the inventors observed that most of the available analytical techniques did not allow for good separation of the constituents of the Lubricating oils. The use of GCxGC allows for good separation of the compounds present in the product, particularly the lubricating oil. Applying an algorithm then allows the analyzed product to be classified into a plurality of classes. It is then possible to determine whether it is recycled oil or conventional oil, including when the analyzed product is a formulated (additized) oil.

[0010] This disclosure also relates to a measurement system (10) comprising: - a chromatography device (12) comprising: + a comprehensive two-dimensional gas chromatography module (13) comprising a first column A and a second column B, capable of separating different compounds of the product according to their volatility and polarity; + a flame ionization detector (14), capable of measuring the intensity of the ionization electric current generated for each compound included in the product sample, - an electronic device (20) for classifying a product into a plurality of classes, the electronic classification device (20) comprising: + a determination module (22) configured to acquire a table describing the intensity of ionization electric current generated for each compound included in the product as a function of the retention times in columns A and B, from a measurement carried out by the chromatography device (12) on the product; + an assignment module (24) configured to assign a class to the product from among the plurality of classes, by applying a multivariate statistical algorithm, said algorithm being trained on tables obtained from reference products.

[0011] This disclosure also relates to a computer program containing instructions for implementing all or part of a process as defined in this disclosure when that program is executed by a processor. This disclosure also relates to a non-transient, computer-readable recording medium on which such a program is recorded.

[0012] The features described in the following paragraphs may optionally be implemented independently of each other or in combination with each other. Brief description of the drawings

[0013] Other features, details and advantages will become apparent upon reading the detailed description below and analyzing the accompanying drawings, in which: Fig. 1

[0014] [Fig.1] is a schematic representation of a measurement system according to the present disclosure, comprising a chromatography device and an electronic characterization device. Fig. 2

[0015] [Fig.2] is a flowchart of the process according to the present disclosure. Fig. 3

[0016] [Fig.3] shows a two-dimensional chromatogram as a function of the retention times of columns A and B obtained from a measurement carried out by the chromatography device (12) on the product. Fig. 4

[0017] [Fig.4] shows a diagram of the different classifications obtained after training the algorithm, in particular according to the example. [Fig. 5]

[0018] [Fig. 5] shows a diagram of the different classifications obtained for the oils tested in example 1. Detailed description

[0019] The present disclosure thus relates to a method of classifying a product among a plurality of classes, via a chromatography device (12), the chromatography device (12) comprising a comprehensive two-dimensional gas chromatography module (13) comprising a first column A and a second column B, and capable of separating different compounds of the product according to their volatility and polarity, the chromatography device (12) further comprising a flame ionization detector (14) capable of measuring an intensity of ionization electric current generated for each compound included in the product, the chromatography device (12) being calibrated with at least one calibration product, allowing the retention time of the different compounds present in the product to be corrected; the process being implemented by an electronic classification device (20) and comprising the following steps: a. determine (100) a table describing the intensity of ionization electric current generated for each compound included in the product as a function of the corrected retention times in columns A and B, from a measurement made by the chromatography device (12) on the product; b. assign a class to the product (120), from among a plurality of classes, by applying a multivariate statistical algorithm to the table, said algorithm being trained on tables obtained from reference products.

[0020] Advantageously, the product is a lubricating oil. This lubricating oil may comprise compounds having a chain length of less than or equal to 40 carbon atoms.

[0021] Lubricating oil (or "lubricant(s)", or "lubricating composition(s)", or "lubricating oil(s)" or "oil(s)") refers to any composition that can be used for the lubrication of moving parts, particularly metal parts, of a mechanical system, such as bearings, gears, or motors. Lubricating oils are generally composed of a base oil and additives; they are then referred to as formulated oils. Lubricating oils may comprise one or more base oils conventionally used in the field of lubricants, such as mineral, synthetic, or natural, animal or vegetable oils, or mixtures thereof.

[0022] These base oils may in particular be oils of mineral or synthetic origin belonging to groups I to V according to the classes defined in the API classification (or their equivalents according to the ATIEL classification) and presented in the following table or their mixtures.

[0023] [Tables 1] Saturates content Sulfur content Viscosity index (VI) Group I Mineral rain < 90% > 0.03% 80 <VI < 120 Groupement II Huiles hydrocraquées >90% <0.03% 80 <VI< 120 Groupement III Huiles hydrocraquées ou hydro-isomérisées > 90% <0.03% >120 Group IV Polyalphaolefins (PAO) Group V Esters and other bases not included in groups I to IV

[0024] During their use, lubricating oils are subjected to stresses that cause their degradation and lead to an increase in the level of undesirable elements, which may originate from the degradation of the base oil itself or of the additives generally present in lubricating compositions, from external pollutants such as dust, or from elements resulting from the wear of parts. which the oil comes into contact with during its use, or even fuel fractions from the engine.

[0025] In consideration of current environmental protection and resource conservation issues, methods have been developed to re-refine, recycle or recondition used lubricating oils.

[0026] It is to the inventors' credit that they discovered that recycled used oils contained specific markers not present in the original oils. This is particularly true of polyalphaolefins (PAOs) and / or normal paraffins (n-paraffins), which are present in recycled used oils at much higher concentrations compared to the original oils.

[0027] According to one embodiment, the present process takes into account the presence or absence of these markers to classify the product.

[0028] Since these markers are present in base oils, the lubricating oil that can be classified under the process of the invention can be an unformulated oil (i.e., a base oil such as CBO or RBO) or a formulated oil (i.e., one with additives for use as a lubricant). The present classification process therefore makes it possible to analyze a base oil and classify it as recycled or not, but also to analyze a formulated lubricating oil and classify it as using a recycled base oil or not. Advantageously, the process is implemented to classify unformulated or formulated oils comprising Group I, II, and / or III base oils.

[0029] Thus, advantageously, the plurality of classes includes at least one original product class and one recycled product class.

[0030] In the process according to the invention, when the tested product consists mainly of an original oil, it is classified in the "original product class," and when the tested product consists mainly of a recycled oil, it is classified in the "recycled product class." A tested oil classified as a recycled product therefore consists mainly of recycled oil and may also include a minor component of original oil.

[0031] When the tested product is an unformulated oil, it is possible to couple the present process with another classification process allowing classification of unformulated oils, using an LC-MS spectrometry module.

[0032] When the tested product is an unformulated oil, it is possible to couple the present process with another classification process allowing classification of unformulated oils, using a spectrometry module, for example a mass spectrometry module with ion cyclotron resonance mass spectrometer, or an Orbitrap type mass spectrometer.

[0033] The chromatography device (12) includes a comprehensive two-dimensional gas chromatography module (13) comprising a first column A and a second column B. The comprehensive two-dimensional gas chromatography modules (2DGC or GCxGC) that can be used in the context of this disclosure are those described in the literature.

[0034] These modules generally comprise an injection module, a vaporization module, a first column A, a modulator, and a second column B. They allow for two-dimensional separation of complex mixtures, because the product is subjected to two separations, resulting in a two-dimensional chromatogram as a function of the retention times of columns A and B and a table describing the intensity of the ionization electric current generated for each compound in the product as a function of the corrected retention times in columns A and B. An example of a two-dimensional chromatogram obtained is shown in [Fig. 3].

[0035] According to one embodiment, the first column A and the second column B are polydimethylsiloxane columns partially functionalized with phenyl groups. The percentage of phenyl functionalization can be between 2 and 50%. In a particular embodiment, the percentage of phenyl functionalization in column A is greater than the percentage of phenyl functionalization in column B. Advantageously, the length of column A is greater than that of column B. The diameter of the two columns A and B can be equivalent. Both columns can have a film thickness of 0.1 µm suitable for separating low-volatility samples. In one embodiment, the temperature gradient applied to the furnace is 2°C / min up to 400°C.

[0036] A quantity of product is injected into the first column A to obtain a first separation, then via the modulator, into the second column B to obtain a second separation. The product can be injected directly without pretreatment, particularly in the case of lubricating oil analysis.

[0037] The GCxGC device is coupled to a flame ionization detector, or FID (14). This detector is capable of measuring the intensity of the ionization current generated for each compound in the product. The flame ionization detector is located at the outlet of the second column.

[0038] Following analysis by the flame ionization detector, a table describing the intensity of the ionization electric current generated for each compound included in the product, as a function of the corrected retention times in columns A and B, is determined (100). The table is therefore derived from a two-dimensional chromatogram obtained from a measurement performed by the chromatography device (12) on the product.

[0039] In addition, during this initial step (100), an external calibration is performed to correct the retention time of the various compounds present in the product. This is carried out by injecting at least one calibration product. If the product to be classified is a lubricating oil, the calibration product can be a lubricating oil, preferably recycled. In one embodiment, the calibration product comprises at least one marker, preferably at least two markers. The marker can be selected from n-paraffins, polyalphaolefins, and mixtures thereof. The correction of the retention times can be performed by software.

[0040] Following this initial step (100), the classification device (20) proceeds to a subsequent step (120), during which it assigns, via its assignment module (28) to the product, a respective class among the plurality of classes, by applying a multivariate statistical algorithm to the table, said algorithm being trained on tables obtained from reference products.

[0041] The multivariate statistical algorithm used in the allocation step (130) may be a multivariate statistical algorithm based on partial least squares regression; the multivariate statistical algorithm preferably being chosen from the group consisting of: a partial least squares regression algorithm, and a partial least squares regression algorithm with discriminant analysis. The algorithm is typically a partial least squares regression algorithm, such as the PLS algorithm or the PLS-DA algorithm.

[0042] The multivariate statistical algorithm used in the allocation step (130) is trained on tables obtained from reference products. In one embodiment, at least one reference product used to train the algorithm includes at least one marker and at least one reference product used to train the algorithm does not include any marker, and at least one class is associated with the presence of said at least one marker, and at least one class is associated with the absence of a marker. The marker can be selected from n-paraffins, polyalphaolefins, and mixtures thereof.

[0043] In the case where the product to be classified is a lubricating oil, the reference products used to train the algorithm may be recycled oils and original oils. Preferably, the reference products used to train the algorithm include at least one recycled oil that may contain at least one marker and at least one original oil that may not contain any marker. The reference products used to train the algorithm may be formulated (additized) or unformulated oils. The number of reference products used to train the algorithm may be between 10 and 500, for example, between 50 and 150.

[0044] The classification thus obtained by the classification device (20) is then particularly effective, as shown – by way of example – by a diagram visible in [Fig.4]. In the example of [Fig.4], the products analyzed are lubricating oils.

[0045] In [Fig. 4], the diagram shows an example of the classification obtained, with the products associated with the first class Cl corresponding to ordinates located above the dotted line, all greater than 0.4; and products associated with the second class C2 corresponding to abscissas located below the line, all less than 0.4.

[0046] The person skilled in the art will then understand that, on this diagram, the classification carried out is quite distinct, with a substantial difference between the abscissas of the two groups, that is to say, of the products of the two distinct classes.

[0047] In the case of [Fig.4], class Cl corresponds to original oils and class C2 to recycled oils.

[0048] The accuracy of the prediction of the classification process can be at least 80%, preferably at least 90%, in particular at least 95%, preferably at least 96% or even at least 97%.

[0049] This disclosure also relates to a method for differentiating recycled oil from conventional oil using the classification process of this disclosure.

[0050] This disclosure also relates to a method for differentiating a lubricating oil formulated from recycled oils from a lubricating oil formulated from conventional oils using the classification process of this disclosure. Measurement system (10)

[0051] This disclosure also relates to a measurement system (10) comprising: - a chromatography device (12) comprising: + a comprehensive two-dimensional gas chromatography module (13) comprising a first column A and a second column B, capable of separating different compounds of the product according to their volatility and polarity; + a flame ionization detector (14), capable of measuring the intensity of the ionization electric current generated for each compound included in the product sample, - an electronic device (20) for classifying a product into a plurality of classes, the electronic classification device (20) comprising: + a determination module (22) configured to acquire a table describing the intensity of the ionization electric current generated for each compound included in the product as a function of the retention times in columns A and B, from a measurement carried out by the chromatography device (12) on the product + an assignment module (24) configured to assign a class to the product from among the plurality of classes, by applying a multivariate statistical algorithm, said algorithm being trained on tables obtained from reference products.

[0052] This measurement system makes it possible to implement the process of the present disclosure. It is presented in [Fig. 1].

[0053] In the example of [Fig.1], the electronic classification device (20) includes an information processing unit (30) formed for example of a memory (32) and a processor (34) associated with the memory (32).

[0054] In the example of [Fig. 1], the determination module (22) and the assignment module (24) are each implemented as a software program, or a software component, executable by the processor (34). The memory (32) of the electronic classification device (20) is then capable of storing determination software, which acquires a table describing the intensity of the ionization current generated for each compound in the product as a function of the retention times in columns A and B, based on a measurement performed by the chromatography device (12) on the product; and assignment software, which assigns each product in the set to a respective class by applying the multivariate statistical algorithm to the table. The processor (34) is then capable of executing each of the software programs, namely the determination software and the assignment software.

[0055] The electronic classification device (20) may not be directly connected to the chromatography device. In this case, the data obtained from the measurements performed by the chromatography device (12) on the product are collected and then transmitted to the determination module (22).

[0056] In an alternative not shown, the determination module (22), and the allocation module (24) are each implemented in the form of a programmable logic component, such as an FPGA (Field Programmable Gate Away), or in the form of a dedicated integrated circuit, such as an ASIC (Application Specified Integrated Circuit).

[0057] When the electronic classification device (20) is implemented in the form of one or more software programs, i.e., in the form of a computer program, it is also capable of being stored on a computer-readable medium, not shown. The computer-readable medium is, for example, a medium capable of storing electronic instructions and being connected to a bus of a computer system. By way of example, the readable medium is an optical disc, a magneto-optical disc, a ROM, a RAM, any type of non-volatile memory (e.g., EPROM, EEPROM, FLASH, NVRAM), a magnetic card, or an optical card. A computer program containing software instructions is then stored on the readable medium.

[0058] By way of example, the table is in the form of a matrix consisting of rows and columns, and each row corresponds to a retention time in column A, each column corresponding to a retention time in column B, and each cell of the matrix then containing the intensity value for the compound corresponding to the column considered and for the product corresponding to the row considered.

[0059] The assignment module (24) is then configured to assign the product a respective class from among the plurality of classes, by applying the multivariate statistical algorithm to the table.

[0060] Thus, the measurement system (10) according to the invention makes it possible to efficiently assign to the product a class among the plurality of classes.

[0061] The application of the multivariate statistical algorithm makes it possible to carry out a precise classification of each of the products tested. Examples

[0062] In this example, the process of the present disclosure is used to determine whether lubricating oils are recycled oils or conventional oils.

[0063] The device used comprises: - a chromatography setup comprising a comprehensive two-dimensional gas chromatography (GCxGC) module coupled to a flame ionization detector (FID). The GCxGC module includes a first column A and a second, shorter column B. These are polydimethylsiloxane-based columns partially functionalized with phenyl groups. The percentage of phenyl functionalization in column A is greater than the percentage of phenyl functionalization in column B. Both columns have a film thickness of 0.1 µm, suitable for separating low-volatility samples. The temperature gradient applied to the furnace is 2°C / min up to 400°C; - an electronic classification device using a PLS-DA algorithm, which has already been trained on commercial references of recycled and conventional lubricating oils. The diagram of the different classes obtained after training is presented in [Fig.4], with the products associated with the conventional oil class corresponding to ordinates located above the dashed line, all greater than 0.4; and the products associated with the recycled oil class corresponding to abscissas located below the line, all less than 0.4.

[0064] The products analyzed are commercially available, unformulated base oils: - 4 conventional base oils: EHC 65, Adbase 4, Nexbase 3030 and Yubase 3 - 2 recycled base oils: Tidrabase Pre N100 and Norbo 52

[0065] The products are not treated before injection, they are analyzed by the GCxGC - FID device and the retention times are corrected, following the injection of a calibration solution, which makes it possible to obtain for each oil a two-dimensional chromatogram as a function of the retention times of columns A and B.

[0066] The data obtained are then transmitted to the classification device and the algorithm is applied. A classification is then obtained for each of the 6 products. The results are presented in [Fig. 5].

[0067] These results show that the process of the present disclosure makes it possible to assign the correct class to each of the 6 products analyzed. Indeed, the 4 conventional base oils appear above the 0.4 dotted line, they are therefore classified as conventional, and the 2 recycled base oils appear below the 0.4 dotted line, they are therefore classified as recycled. Embodiments

[0068] This disclosure relates to different embodiments as presented below: Embodiment 1: A method for classifying a product into a plurality of classes, via a chromatography device (12), the chromatography device (12) comprising a comprehensive two-dimensional gas chromatography module (13) comprising a first column A and a second column B, and capable of separating different compounds of the product according to their volatility and polarity, the chromatography device (12) further comprising a flame ionization detector (14) capable of measuring an intensity of ionization electric current generated for each compound included in the product, the chromatography device (12) being calibrated with at least one calibration product, allowing the retention time of the different compounds present in the product to be corrected; the process being implemented by an electronic classification device (20) and comprising the following steps: a. determine (100) a table describing the intensity of ionization electric current generated for each compound included in the product as a function of the corrected retention times in columns A and B, from a measurement made by the chromatography device (12) on the product; b. assign a class to the product (120), from among a plurality of classes, by applying a multivariate statistical algorithm to the table, said algorithm being trained on tables obtained from reference products.

[0069] Embodiment 2: Process according to embodiment 1, wherein the product is a lubricating oil.

[0070] Embodiment 3: Process according to embodiment 2, in which the lubricating oil comprises compounds having a chain length less than or equal to 40 carbon atoms.

[0071] Embodiment 4: A method according to any one of the preceding embodiments, wherein the plurality of classes comprises at least one original product class and one recycled product class.

[0072] Embodiment 5: Method according to any one of the preceding embodiments, in which the table is derived from a two-dimensional chromatogram obtained from a measurement carried out by the chromatography device (12) on the product.

[0073] Embodiment 6: Method according to any one of the preceding embodiments, in which the comprehensive two-dimensional gas chromatography module (13) comprises an injection module, a vaporization module, a first column A, a modulator, and a second column B.

[0074] Embodiment 7: Process according to the preceding embodiment, in which the first column A and the second column B are columns based on polydimethylsiloxane functionalized in part with phenyl groups.

[0075] Embodiment 8: A method according to any one of the preceding embodiments, wherein at least one reference product used to train the algorithm includes at least one marker and at least one reference product used to train the algorithm does not include any marker, and at least one class is associated with the presence of said at least one marker, and at least one class is associated with the absence of a marker.

[0076] Embodiment 9: A method according to any one of the preceding embodiments, in which the calibration product comprises at least one marker, preferably at least two markers.

[0077] Embodiment 10: Method according to embodiment 8 or 9, wherein the marker is selected from n-paraffins, polyalphaolefins and their mixture.

[0078] Embodiment 11: A method according to any one of the preceding embodiments, wherein the multivariate statistical algorithm is a multivariate statistical algorithm by partial least squares regression; the multivariate statistical algorithm being preferably chosen from the group consisting of: a partial least squares regression algorithm, and a partial least squares regression algorithm with discriminant analysis.

[0079] Embodiment 12: Measuring system (10) comprising: - a chromatography device (12) comprising: + a two-dimensional integral gas chromatography module (13) comprising a first column A and a second column B, capable of separating different compounds of the product according to their volatility + a flame ionization detector (14), capable of measuring the intensity of the ionization electric current generated for each compound included in the product sample, - an electronic device (20) for classifying a product into a plurality of classes, the electronic classification device (20) comprising: + a determination module (22) configured to acquire a table describing the intensity of ionization electric current generated for each compound included in the product as a function of the retention times in columns A and B, from a measurement carried out by the chromatography device (12) on the product + an assignment module (24) configured to assign a class to the product from among the plurality of classes, by applying a multivariate statistical algorithm, said algorithm being trained on tables obtained from reference products.

[0080] Embodiment 13: Computer program comprising instructions for implementing the process according to one of the embodiments 1 to 11 when this program is executed by a processor.

[0081] Embodiment 14: Non-transient computer-readable recording medium on which a program is recorded for the implementation of the process according to one of embodiments 1 to 11 when this program is executed by a processor.

Claims

Demands

1. A method for classifying a product into a plurality of classes, via a chromatography device (12), the chromatography device (12) comprising a comprehensive two-dimensional gas chromatography module (13) comprising a first column A and a second column B, and capable of separating different compounds of the product according to their volatility and polarity, the chromatography device (12) further comprising a flame ionization detector (14) capable of measuring an ionization current intensity generated for each compound included in the product, the chromatography device (12) being calibrated with at least one calibration product, enabling correction of the retention time of the different compounds present in the product; the method being implemented by an electronic classification device (20) and comprising the following steps: a.determine (100) a table describing the intensity of ionization electric current generated for each compound included in the product as a function of the corrected retention times in columns A and B, from a measurement made by the chromatography device (12) on the product; b. assign a class to the product (120), from among a plurality of classes, by applying a multivariate statistical algorithm to the table, said algorithm being trained on tables obtained from reference products.

2. A method according to claim 1, wherein the product is a lubricating oil.

3. A method according to the preceding claim, wherein the lubricating oil comprises compounds having a chain length less than or equal to 40 carbon atoms.

4. A method according to any one of the preceding claims, wherein the plurality of classes comprises at least one original product class and one recycled product class.

5. A method according to any one of the preceding claims, wherein the table is derived from a two-dimensional chromatogram obtained from a measurement carried out by the chromatography device (12) on the product.

6. A method according to any one of the claims, wherein the comprehensive two-dimensional gas chromatography module (13) comprises an injection module, a vaporization module, a first column A, a modulator, and a second column B.

7. A method according to the preceding claim, wherein the first column A and the second column B are columns based on polydimethylsiloxane functionalized in part with phenyl groups.

8. A method according to any one of the preceding claims, wherein at least one reference product used to train the algorithm comprises at least one marker and at least one reference product used to train the algorithm does not comprise any marker, and at least one class is associated with the presence of said at least one marker, and at least one class is associated with the absence of the marker.

9. A method according to any one of the preceding claims, wherein the calibration product comprises at least one marker, preferably at least two markers.

10. A method according to any one of claims 8 or 9, wherein the marker is selected from n-paraffins, polyalphaolefins and mixtures thereof.

11. A method according to any one of the preceding claims, wherein the multivariate statistical algorithm is a multivariate statistical algorithm by partial least squares regression; the multivariate statistical algorithm being preferably chosen from the group consisting of: a partial least squares regression algorithm, and a partial least squares regression algorithm with discriminant analysis.

12. A measurement system (10) comprising: - a chromatography device (12) comprising: + a two-dimensional integral gas chromatography module (13) comprising a first column A and a second column B, capable of separating different compounds of the product according to their volatility + a flame ionization detector (14), capable of measuring an intensity of ionization electric current generated for each compound included in the product sample, - an electronic classification device (20) for a product among a plurality of classes, the electronic classification device (20) comprising: + a determination module (22) configured to acquire a table describing the intensity of ionization electric current generated for each compound included in the product as a function of the retention times in columns A and B, from a measurement carried out by the chromatography device (12) on the product + an assignment module (24) configured to assign a class to the product among the plurality of classes, by applying a multivariate statistical algorithm, said algorithm being trained on tables obtained from reference products.

13. Computer program comprising instructions for carrying out the method according to any one of claims 1 to 11 when this program is executed by a processor.

14. Non-transient computer-readable recording medium on which is recorded a program for implementing the method according to any one of claims 1 to 11 when this program is executed by a processor.

Images