Pet discrimination method

Abstract

Provided is a PET discrimination method that is an objective method that offers comparatively simple differentiating operation and that makes it possible to differentiate between homogeneous polyester resins (including fiber). Infrared or near-infrared absorption spectra of respective comparison resins, of a group A of a plurality of comparison resins that are recycled PET resins and a group B of a plurality of comparisons resins that are virgin PET resins, are found. Spectral data in a predetermined wavenumber range is extracted from the absorption spectra, and an analysis data group obtained by multivariate analysis of the spectral data is accumulated as a database. Next, an unknown resin is taken to be a resin under test, and test data obtained by extracting spectral data in the predetermined wavenumber range from the resulting absorption spectra is found and cross-checked against the analysis data group of the database by multivariate analysis, to differentiate whether the resin under test is recycled PET resin or virgin PET resin.

Description

PET identification method

[0001] The present invention relates to a PET discrimination method for discriminating the type of polyethylene terephthalate resin (PET resin) used in fibers, films, and other resin molded products. In particular, the present invention relates to a PET discrimination method for discriminating between PET fibers made of PET resin synthesized by conventional methods and PET fibers recycled from used PET bottles.

[0002] Many resin products are available on the market for a wide range of applications. For example, synthetic resins such as nylon, polyester, and acrylic are used in textile products and films. Among these, polyester resins are widely used in clothing fibers, films, PET bottles, and the like. Polyester fibers are the most widely produced clothing fibers. Polyester fibers used in clothing are generally made of polyethylene terephthalate (PET), and have been widely used for a long time due to their physical properties suitable for clothing.

[0003] In addition to PET, other polyester resins include polybutylene terephthalate (PBT), polytrimethylene terephthalate (PTT), and polyethylene naphthalate (PEN). PBT and PTT are also used as fibers, and PTT is an environmentally friendly plant-derived material. PET, PBT, and PEN are also used as engineering plastics.

[0004] With the recent rise in interest in environmental issues, there has been a growing demand for environmentally conscious materials for clothing. However, there is still a large demand for synthetic fibers, and among them, polyester fiber (PET fiber) is an essential fiber for clothing. On the other hand, PTT, an environmentally friendly material, is also used as a fiber due to its good stretchability. However, PTT fiber cannot replace PET fiber as a clothing material due to its basic physical properties and cost.

[0005] Meanwhile, PET is being recycled and reused as PET resin. Methods for recycling PET and reusing it as PET resin include chemical recycling and material recycling. Chemical recycling of PET involves returning used PET products recovered from use back to their synthetic raw materials, terephthalic acid or dimethyl terephthalate (DMT), which are then polycondensed with ethylene glycol to synthesize PET again. This method has the advantage that PET products that have already been used can be reused even if they contain other components such as dyes and pigments. However, there are problems with the equipment and costs involved in performing depolymerization and re-polycondensation.

[0006] In addition, PET material recycling involves using PET products collected after use as PET. PET is a thermoplastic resin that is heated and melted to form pellets of plastic raw material. These pellets are then processed into PET fiber or PET film using conventional methods. This method does not require depolymerization and re-polycondensation, so there are fewer problems in terms of equipment and cost. However, there is a problem in that the use of PET products that have been used is limited if they contain other ingredients such as dyes or pigments.

[0007] Therefore, in the material recycling of PET, the raw material is collected and reused from PET bottles, which are widely distributed on the market among PET products that have been used once. Collected PET bottles are made of PET resin and do not contain other components such as dyes or pigments. In addition, a collection system for PET bottles has been established in the market. Currently, recycled PET is being promoted from both chemical recycling and material recycling, but in the future, it is expected that the distribution of recycled PET made from material recycled PET from collected PET bottles will increase.

[0008] Therefore, in this invention, the terms "recycled PET resin" and "recycled PET fiber" are used to refer only to PET that has been material recycled from recovered PET bottles, and do not include chemically recycled PET. In contrast, the terms "virgin PET resin" and "virgin PET fiber" are used to refer to regular PET that has been newly polycondensed from monomers. Hereinafter, the present invention will be described using the terms "recycled PET resin" and "virgin PET resin," or "recycled PET fiber" and "virgin PET fiber."

[0009] Below, we will explain PET resin using PET fiber as a clothing fiber as an example. In recent years, as interest in sustainability has increased in the textile and apparel industry, the keywords traceability and transparency have become important. This is not limited to natural fibers, but will also become an important element for synthetic fibers in the future.

[0010] On the other hand, due to the fragmented industrial structure of the textile industry, consisting of upstream, midstream, and downstream processes, it is difficult to track information on the manufacturing process of many clothing products. While the spread of recycled PET fiber is expected, traceability is said to be difficult due to the complex manufacturing process from collection to spinning. Furthermore, under Japanese law, both recycled PET fiber and virgin PET fiber are labeled as "polyester," and there is no obligation to label them at the time of sale, making it difficult for environmentally conscious consumers to distinguish between recycled and virgin PET fiber.

[0011] In today's world where textile production and consumption are expanding globally, textile inspection organizations in each country perform fiber identification to ensure safety and reliability in the import and export of textile products. In Japan, for example, these inspection organizations perform identification based on JIS L 1030-1 (Test methods for blending ratios in textile products - Part 1: Fiber identification). Fiber identification methods include combustion tests, tests to confirm chlorine in fibers, tests to confirm nitrogen in fibers, microscopic tests, color tests using iodine-potassium iodide solutions, xanthoprotein reaction tests, and infrared absorption spectrum measurements.

[0012] Each of these test methods is effective, and combining them can distinguish between many fibers. However, among natural fibers, fibers with the same chemical composition (called "similar fibers"), such as cellulosic fibers like cotton, various types of hemp, various types of rayon, cuprammonium rayon, and solvent-spun cellulose fibers, and animal hair fibers like cashmere, wool, yak, mohair, angora, alpaca, vicuna, camel, and llama, are distinguished as identical or similar fibers by the chemical test methods mentioned above. Therefore, these chemical test methods cannot clearly distinguish between similar fibers. The same is true for synthetic fibers.

[0013] The above test methods also include infrared absorption spectrum measurement tests. However, this measurement test is primarily intended to distinguish fibers with different chemical compositions (these are called "different fibers"), and is not considered to be able to clearly distinguish between fibers of the same type. In particular, it has been difficult to distinguish between cellulose-based fibers and animal hair-based fibers, which are different types of natural fibers, using infrared spectroscopy. In response to this, the present inventors have investigated methods for analyzing infrared absorption spectra and have proposed a fiber identification method for different types of natural fibers, as described in Patent Document 1 below.

[0014] On the other hand, with regard to chemical fibers, it is easy to distinguish between dissimilar fibers such as rayon, nylon, polyester, and acrylic by measuring their infrared absorption spectra. Chemical fibers have artificially regenerated and synthesized chemical compositions and are highly uniform, and some dissimilar fibers of the same type can be analyzed to some extent from their infrared absorption spectra, compared to natural fibers. For example, Non-Patent Document 1 below states that PET fibers, PBT fibers, and PEN fibers, which are dissimilar polyester fibers, can be distinguished by focusing on the infrared absorption peaks of the different monomers because they have different combinations of monomers in the ester reaction.

[0015] In contrast, recycled PET fibers and virgin PET fibers, which are both PET fibers, are of the same family and type (these are called "homogeneous fibers"), and therefore it is difficult to distinguish them by infrared spectroscopy. As a method for distinguishing between the same family and type of polyester fibers, a method for distinguishing between these polyester fibers by instrumental analysis other than infrared spectroscopy has been established, for example, in the Chinese standard described in Non-Patent Document 2 below.

[0016] Patent No. 6798885

[0017] Analysis of foreign matter spectra 7 Polyester: PerkinElmer Japan Co., Ltd. (https: / / www.perkinelmer.co.jp / tabid / 2585 / Default.aspx) Chinese standard: GB / T39026-2020 "Identification method for recycled PET fibers"

[0018] The method of Non-Patent Document 2 uses high performance liquid chromatography (HPLC) to analyze trace components contained in PET bottles, but has the problem that the pretreatment by transesterification reaction and the equipment used are special, and the analytical operation is complicated.

[0019] Therefore, the present invention addresses the above problems and aims to provide a PET discrimination method that is relatively simple and objective, and can distinguish between polyester-based resins of the same type and type (including fibers).

[0020] In order to solve the above problems, the present inventors conducted extensive research and discovered that it is possible to distinguish between the same type of polyester resin (including fibers) through simple operation and analysis even using an infrared absorption spectrometer that is widely used in industry, and this led to the completion of the present invention.

[0021] That is, according to claim 1, the PET discrimination method of the present invention is a PET discrimination method for discriminating between recycled PET resin and virgin PET resin in polyethylene terephthalate resin (PET resin), comprising the steps of: classifying a plurality of comparative resins known to be recycled PET resins as group A, and a plurality of comparative resins known to be virgin PET resins as group B, irradiating the comparative resins of each group with infrared or near-infrared light to obtain their respective absorption spectra; extracting spectral data in a predetermined wavenumber range from these absorption spectra, and performing multivariate analysis on the spectral data to obtain a group of analytical data as a database; classifying a resin whose identity as recycled PET resin or virgin PET resin is unknown as a test resin, and obtaining test data by extracting spectral data in the same predetermined wavenumber range as that of the comparative resin to which the test resin is to be compared, from the absorption spectrum obtained from the test resin in the same manner as in the database creation step; The method also includes a discrimination process in which the test data is compared with the analytical data group in the database by the multivariate analysis, and the consistency between the test data and the analytical data group in the database is used as an index to determine whether the test resin is a recycled PET resin or a virgin PET resin.

[0022] According to a second aspect of the present invention, there is provided a PET discrimination method according to the first aspect, wherein the comparative resin and the test resin of each group are subjected to a PET discrimination at a wave number of 4000 cm －１ ~600cm －１ the absorption spectrum is obtained by irradiating infrared light within a range of 1000 to 1500 nm, the spectral data in the predetermined wave number range is extracted to obtain the database and the test data, and it is determined whether the test resin is a recycled PET resin or a virgin PET resin.

[0023] According to a third aspect of the present invention, there is provided the PET discrimination method of the second aspect, wherein the predetermined wavenumber range is a wavenumber range of 1400 cm －１ ~600cm －１ The spectral data in these wavenumber ranges is used for analysis.

[0024] According to a fourth aspect of the present invention, there is provided the PET discrimination method according to any one of the first to third aspects, wherein the multivariate analysis is partial least squares discriminant analysis (PLS-DA).

[0025] According to a fifth aspect of the present invention, there is provided the PET distinguishing method of the fourth aspect, wherein the PET resin is a fibrous PET fiber.

[0026] Furthermore, according to the description of claim 6, the present invention is the PET discrimination method described in claim 4, characterized in that the method for determining the absorption spectra of the comparative resin and the test resin is the ATR method (attenuated total reflection method).

[0027] According to a seventh aspect of the present invention, there is provided the PET distinguishing method of the sixth aspect, wherein the PET resin is a fibrous PET fiber.

[0028] According to the above configuration, the present invention is a PET discrimination method for discriminating between recycled PET resin and virgin PET resin in polyethylene terephthalate resin (PET resin), a type of polyester resin. Similar resins of the same type and chemical composition can be distinguished from each other by infrared or near-infrared absorption spectrum. According to the above configuration, the present invention includes a database creation step and a discrimination step.

[0029] In the database creation process, a plurality of comparative resins known to be recycled PET resins are grouped as Group A, and a plurality of comparative resins known to be virgin PET resins are grouped as Group B. Next, the comparative resins in each group are irradiated with infrared or near-infrared light to obtain their respective absorption spectra. Spectral data in a predetermined wavenumber range is extracted from these absorption spectra, and the spectral data is subjected to multivariate analysis, resulting in a group of analytical data that is stored as a database.

[0030] Next, in the discrimination step, a resin whose identity as a recycled PET resin or a virgin PET resin is unknown is used as a test resin, and test data is obtained by extracting spectral data in a predetermined wavenumber range similar to that of a comparison resin to which the test resin is to be compared from the absorption spectrum obtained from the test resin in the same manner as in the database creation step. Next, the test data is compared with a group of analytical data in the database by multivariate analysis, and whether the test resin is a recycled PET resin or a virgin PET resin is determined using the agreement between the test data and the group of analytical data in the database as an index.

[0031] Furthermore, according to the above configuration, the comparative resin and the test resin of each group were measured at a wave number of 4000 cm －１ ~600cm －１ The absorption spectrum may be obtained by irradiating infrared light within the range of 1400 cm. －１ ~600cm －１ or a range close thereto. Furthermore, according to the above configuration, the multivariate analysis is preferably partial least squares discriminant analysis (PLS-DA). Furthermore, the method for determining the absorption spectra of the comparative resin and the test resin may be the ATR method (attenuated total reflection method). Furthermore, the PET resin may be a fibrous PET fiber.

[0032] Therefore, according to the present invention, a PET discrimination method can be provided that is capable of discriminating between polyester-based resins of the same type (including fibers) using a relatively simple and objective discrimination operation.

[0033] 1 is a diagram in which the infrared absorption spectra of recycled PET fiber and virgin PET fiber are superimposed. It is a diagram showing a spectrum (differential spectrum) obtained by second-order differentiation of the infrared absorption spectrum of FIG. 1. It is a diagram showing the differential spectrum of FIG. 2 with wavenumber selection. It is a diagram showing the differential spectrum of FIG. 3 with wavenumber selection subjected to SNV conversion. It is a diagram showing the centered SNV-converted differential spectrum of FIG. 4. It is a diagram showing a discrimination model obtained by PLS-DA. It is a diagram showing discrimination of recycled PET fiber using the discrimination model obtained in FIG. 6. It is a diagram showing discrimination of virgin PET fiber using the discrimination model obtained in FIG. 6.

[0034] In explaining the present invention, PET fiber will be used as a specific example of PET resin. Therefore, in the following explanation, PET resin will be referred to as PET fiber, recycled PET resin will be referred to as recycled PET fiber, and virgin PET resin will be referred to as virgin PET fiber. Note that, although PET resin in fibrous form will be described as PET fiber, the present invention is not limited to fibers and also covers films and other resin molded products.

[0035] In the present invention, polyester fibers are the focus of attention among the fibers generally used in various textile products such as clothing and industrial materials. Polyester fibers include polyethylene terephthalate fibers (PET fibers), polybutylene terephthalate fibers (PBT fibers), and polytrimethylene terephthalate fibers (PTT fibers), which are defined as "homogeneous fibers." However, the present invention is not intended to distinguish between these homogeneous fibers.

[0036] The present invention is directed to distinguishing between PET fibers, and defines these as "same kind of fibers." Distinguishing between same kind of fibers is considered to be extremely difficult. In the present invention, the same kind of fibers of PET fibers are limited to PET fibers that have been material recycled from recovered PET bottles, as described above, and are referred to as "recycled PET fibers." In contrast, ordinary PET fibers newly polycondensed from monomers are referred to as "virgin PET fibers."

[0037] Here, we will explain virgin PET fiber and recycled PET fiber. Virgin PET fiber is produced by spinning PET resin synthesized by polycondensation of dimethyl terephthalate (DMT) and ethylene glycol (DMT method) or polycondensation of terephthalic acid and ethylene glycol (direct polymerization method). In contrast, recycled PET fiber is produced by spinning PET resin recycled from recovered PET bottles. It is said that in the PET resin used in recovered PET bottles, about 2% of the terephthalic acid used in polycondensation of the PET resin is replaced with isophthalic acid to improve the bottle's appearance.

[0038] Even if a small amount of isophthalic acid is added to terephthalic acid, the polymerization monomer for PET fiber, it is difficult to distinguish it as a homogeneous polyester fiber using infrared spectroscopy, and it must be considered a homogeneous fiber. In reality, beverage bottles made from polyester resins containing a small amount of isophthalic acid are distributed as PET bottles. Therefore, in order to distinguish PET resin recycled from recovered PET bottles from conventional PET resin, complicated analysis by HPLC is required, as in the method described in Non-Patent Document 2 above.

[0039] In contrast, in the present invention, the discrimination between polyester fibers of the same type and different types is performed by infrared spectroscopy, which is a relatively simple and objective method. In the present invention, discrimination can be performed not only by infrared spectroscopy but also by near-infrared spectroscopy. Hereinafter, the present invention will be described using infrared spectroscopy.

[0040] The present inventors measured the infrared absorption spectra of recycled PET fibers and virgin PET fibers and compared them. Figure 1 shows the infrared absorption spectra of recycled PET fibers and virgin PET fibers superimposed on each other. In Figure 1, the infrared absorption spectra of recycled PET fibers and virgin PET fibers have overlapping absorption peaks across the entire measured wavenumber range, making it impossible to distinguish between them on their own. Furthermore, the present inventors enlarged each wavenumber range and conducted a detailed study, but were unable to distinguish between them.

[0041] Infrared spectroscopy obtains an infrared absorption spectrum based on the molecular vibrations of the substance being measured. Therefore, the infrared absorption spectrum contains information about the substance's specific functional groups and crystalline structure. Even if a small amount of isophthalic acid is mixed into terephthalic acid, one of the polymerization monomers of the PET resin that is the subject of this invention, these functional groups are the same. In other words, the two carboxylic acids that are the functional groups of phthalic acid (benzenedicarboxylic acid) are the same, but their relative positions are slightly different. Terephthalic acid is the para-isomer of benzenedicarboxylic acid, and isophthalic acid is the meta-isomer of benzenedicarboxylic acid. These are structural isomers.

[0042] The inventors therefore considered the following: When para- and meta-benzenedicarboxylic acids form a linear polymer, differences in the configuration of some of the monomers would cause differences in the conformation of the PET resin, resulting in differences in the crystal structure. They believed that these slight differences in crystal structure could be distinguished by multivariate analysis of multiple infrared absorption spectra.

[0043] First, we will explain the method for obtaining infrared absorption spectra from fibers of known fiber type (hereinafter referred to as "comparison fibers") and fibers to be identified (hereinafter referred to as "test fibers") when creating a database in the present invention. This method is generally called infrared spectroscopy, and involves irradiating the substance to be measured with infrared light and separating the transmitted or reflected light to obtain an infrared absorption spectrum, thereby determining the characteristics of the object. Infrared spectroscopy is used to determine the molecular structure of an object and is an extremely common method for analyzing organic materials.

[0044] In general, in infrared spectroscopy, wave numbers (also referred to as "WN" in the present invention) are often used rather than wavelengths, and the wave number range used in the present invention is 4000 cm －１ ~600cm －１ On the other hand, the wave number range used in near-infrared spectroscopy is 12500 cm －１ ~4000cm －１ Within the range.

[0045] In the present invention, a commonly used Fourier transform infrared spectrophotometer (hereinafter referred to as "FT / IR spectrophotometer") can be used. To obtain the infrared absorption spectra of the comparison fiber and the test fiber, the fiber may be pulverized and formed into a tablet with potassium bromide (KBr) powder, and the transmitted light may be measured by the KBr tablet method. Alternatively, the measurement may be performed by the ATR method (attenuated total reflection method), which can measure reflected light directly from the woven or knitted fabric without destroying it. It is also preferable to perform atmospheric correction, baseline correction, smoothing correction, penetration depth correction, etc., as necessary, using conventional methods for the infrared absorption spectra obtained by the ATR method.

[0046] The PET discrimination method according to the present invention will be described in detail below with reference to an embodiment for PET fibers. Note that the present invention is not limited to the following embodiment. Note that the present invention comprises a "database creation step" and a "discrimination step."

[0047] In the present invention, the infrared absorption spectra of a plurality of comparative fibers (Group A) known to be recycled PET fibers and the infrared absorption spectra of a plurality of comparative fibers (Group B) known to be virgin PET fibers are obtained. Next, spectral data in a predetermined wavenumber range important for analysis is extracted from these absorption spectra. Next, a group of analytical data obtained by multivariate analysis of this spectral data is stored as a database. This process is called the "database creation process."

[0048] Next, a fiber whose identity as a recycled PET fiber or a virgin PET fiber is unknown is used as the test fiber in the same manner, and its infrared absorption spectrum is determined. Next, test data is obtained by extracting spectral data in a predetermined wavenumber range similar to that of the comparison fibers of Groups A and B from the obtained infrared absorption spectrum of the test fiber. The obtained test data is then compared with the analysis data group of the database by the above-mentioned multivariate analysis. In this comparison, the agreement between the test data of the test fiber and the analysis data group of the database of the comparison fiber is used as an index to determine whether the test fiber is a recycled PET fiber or a virgin PET fiber. This process is referred to as the "determination process." Each process is described below.

[0049] (1) Database Creation Process In this embodiment, 100 samples known to be recycled PET fibers were designated as comparative fibers in Group A, and 100 samples known to be virgin PET fibers were designated as comparative fibers in Group B. For these comparative fibers in Group A and Group B, a wave number of 4000 cm was used. －１ ~600cm －１The infrared absorption spectra were obtained within the range of 100 Hz. Figure 1 shows the infrared absorption spectra of the recycled PET fibers and virgin PET fibers superimposed on each other. The infrared absorption spectra in Figure 1 were corrected with ATR. As mentioned above, the infrared absorption spectra of the recycled PET fibers (Group A) and virgin PET fibers (Group B) shown in Figure 1 have overlapping absorption peaks across the entire measured wavenumber range, making it impossible to distinguish between them on their own.

[0050] In this embodiment, an FT / IR spectrophotometer (FT / IR-6800 manufactured by JASCO Corporation) was used to determine the infrared absorption spectrum of the comparative fiber. FT / IR spectrophotometers are inexpensive because they are used in many industries, and the investment in equipment and maintenance costs for identification are reduced. Furthermore, in this embodiment, the ATR method (attenuated total reflection measurement method) was used for the measurement. In many cases, the fibers to be identified are already in the form of woven or knitted textile products, and this method has the advantage of being able to identify the fibers non-destructively while they are still in this woven or knitted state. The type of prism used in the ATR method is not particularly limited, but a diamond prism was used in this embodiment.

[0051] Here, when analyzing the spectral data of the comparative fiber, it is preferable to first preprocess the spectral data. Preprocessing preferably involves, for example, first-order differentiation, second-order differentiation, or higher-order differentiation. These differentiation processes can enhance spectral information, such as by sharpening peaks buried in the spectrum or eliminating background effects. While the method and dimension used in differentiation are not particularly limited, in this embodiment, second-order differentiation using the Savitzky-Golay method (SG method) is preferable. Figure 2 shows a spectrum (differential spectrum) obtained by second-order differentiation of an infrared absorption spectrum.

[0052] Next, spectral data in a predetermined wavenumber range effective for analysis is extracted from the obtained differential spectrum. The method for extracting spectral data is not particularly limited. A wavenumber range corresponding to a specific functional group considered necessary for analysis may be selected for the type and combination of fibers to be distinguished. The inventors selected a wavenumber range related to the functional group of isophthalic acid, which is considered to affect the crystallinity of PET fiber. Spectral data in a wavenumber range considered to contain noise in the analysis may also be actively eliminated. Note that, as a result of extensive research, the inventors have determined that in order to distinguish between recycled PET fiber and virgin PET fiber, a wavenumber range of 1,400 cm is the most suitable. －１ ~600cm －１ It has been found that it is preferable to extract spectral data within or near the range of Fig. 3. Fig. 3 shows the differential spectrum with selected wave numbers.

[0053] Next, in this embodiment, the wavenumber-selected differential spectrum was transformed using standard normal variation (SNV). SNV transformation is a method of normalizing each spectrum so that the average of the constituent data is 0 and the standard deviation is 1. FIG. 4 shows the SNV-transformed wavenumber-selected differential spectrum. Furthermore, mean centering was performed on scattering variations that could not be experimentally resolved. FIG. 5 shows the centered SNV-transformed differential spectrum.

[0054] Next, the extracted spectral data is subjected to multivariate analysis to obtain an analysis data group. The analysis method used for the multivariate analysis is not particularly limited, and any method commonly used in multivariate analysis may be adopted. Generally, multivariate analysis methods include principal component analysis, multiple regression analysis, independent component analysis, factor analysis, discriminant analysis, quantification theory, cluster analysis, and multidimensional scaling. In the present invention, of these, it is preferable to use multiple regression analysis such as principal component analysis (PCA), principal component regression (PCR), or PLS regression, and it is particularly preferable to use partial least squares discriminant analysis (PLS-DA).

[0055] In this embodiment, PLS-DA was used. The creation of a discrimination model using PLS-DA will be specifically described below. Sixty samples of the recycled PET fibers (Group A) and sixty samples of the virgin PET fibers (Group B) prepared as described above were used as comparison fibers. The results were analyzed at a wave number of 1400 cm. －１ ~600cm －１ A discriminant model was created using a database of analytical data obtained by analyzing spectral data within this range using PLS-DA. In this discriminant model, a value of 1 corresponds to group A and a value of -1 corresponds to group B, and if the predicted value on the vertical axis is positive, it is determined to be recycled PET fiber, and if it is negative, it is determined to be virgin PET fiber.

[0056] Figure 6 shows the discrimination model obtained by PLS-DA. The number of factors used to create the discrimination model was 7. In Figure 6, the horizontal axis represents the sample number (comparison fibers 1 to 120), and the vertical axis represents the predicted value of each sample. The predicted values ​​of the comparison fibers were all positive for recycled PET fibers and all negative for virgin PET fibers, allowing for accurate discrimination.

[0057] (2) Discrimination Step In this discrimination step, first, an infrared absorption spectrum is obtained for a test fiber, which is unknown whether it is a recycled PET fiber or a virgin PET fiber. In this embodiment, in order to verify the discrimination model, 18 samples known to be recycled PET fibers and 36 samples known to be virgin PET fibers were used as provisional test fibers for verification. In addition, two chemically recycled PET fibers were added to the provisional test fibers, and it was verified how these fibers could be distinguished.

[0058] The methods for obtaining the infrared absorption spectrum, the methods for making various corrections to the infrared absorption spectrum, and the methods for performing preprocessing such as second-order differential processing, wavenumber range extraction, SNV conversion, and centering processing are the same as those for the comparison fiber described above.

[0059] The results of discrimination using the PLS-DA discrimination model for the test fibers are shown in Figures 7 and 8. Figure 7 shows the discrimination of recycled PET fibers using the discrimination model. In Figure 7, 18 tentative test fibers that were recycled PET fibers were correctly identified as recycled PET fibers. Also, in Figure 7, two tentative test fibers that were chemically recycled PET fibers were identified as virgin PET fibers. Chemically recycled PET fibers are recovered from recycled PET bottles, but are recycled down to polymerized monomers through depolymerization. Chemically recycled PET fibers are obtained by polycondensing recycled polymerized monomers with ethylene glycol, and are considered to be indistinguishable from virgin PET fibers in terms of quality.

[0060] On the other hand, Figure 8 is a diagram showing the discrimination of virgin PET fibers using the discrimination model. In Figure 8, 36 tentative test fibers, which are virgin PET fibers, were correctly discriminated as virgin PET fibers. These 36 tentative test fibers were fibers randomly selected from commercially available products, and included woven fabrics, knitted fabrics, filament yarns, and staple yarns, and some were dyed fibers. However, according to the discrimination model of this embodiment, all of these virgin PET fibers randomly selected from commercially available products were correctly discriminated.

[0061] As described above, the present invention provides a PET discrimination method that is relatively simple and objective, and that can discriminate between polyester-based resins of the same type and kind (including fibers).

[0062] It should be noted that the present invention is not limited to the above-described embodiment, and various modifications such as the following can be made. (1) In the above-described embodiment, the discrimination is made based on the reflectance spectrum obtained by infrared spectroscopic analysis using the ATR method, but the present invention is not limited to this, and the discrimination may be made based on the transmission spectrum obtained by infrared spectroscopic analysis. (2) In the above-described embodiment, the infrared absorption spectrum is obtained by the ATR method, but the present invention is not limited to this, and the discrimination may be made based on other methods, for example, the infrared absorption spectrum may be obtained by finely pulverizing the comparison fiber or test fiber and then using the KBr tablet method. (3) In the above-described embodiment, the discrimination may be made based on the reflectance spectrum obtained by infrared spectroscopic analysis using the ATR method, but the discrimination may be made based on the transmittance spectrum obtained by infrared spectroscopic analysis. －１ In the above embodiment, the infrared absorption spectrum is measured within a range of 1400 cm wavenumber, and spectral data in a predetermined wavenumber range to be used for analysis is extracted from the measured spectrum. However, the present invention is not limited to this, and the infrared absorption spectrum in a predetermined wavenumber range to be used for analysis may be directly measured. －１ ~600cm －１ However, the present invention is not limited to this, and other wavelength ranges may be extracted. (5) In the above embodiment, the spectral data is analyzed by PLS-DA, but the present invention is not limited to this, and the spectral data may be analyzed by principal component analysis (PCA), principal component regression (PCR), PLS regression, or other multivariate analysis. (6) In the above embodiment, the discrimination is performed using an infrared absorption spectrum obtained by infrared spectroscopic analysis, but the present invention is not limited to this, and the discrimination may be performed using a near-infrared absorption spectrum obtained by near-infrared spectroscopic analysis. (7) In the above embodiment, the discrimination is performed for PET fiber as a PET resin product, but the present invention is not limited to this, and the discrimination may be performed for PET film and other PET resin molded products.

[0063] Many resin and textile products are available on the market for a wide range of applications. In addition, as the production and consumption of resin and textile products expand globally, a rapid and accurate method for identifying these products is needed to ensure safety and reliability during import and export. In particular, accurate identification of the same type of polyester resin (including fibers) with nearly identical chemical compositions is needed.

[0064] The present invention provides an accurate discrimination means that meets such market demands, and is an unprecedented, groundbreaking discrimination means that can objectively distinguish between the types of polyester-based resins of the same type (including fibers) and can provide accurate discrimination results even if erroneous or false descriptions are made.

Claims

1. A PET discrimination method for discriminating between recycled PET resin and virgin PET resin in polyethylene terephthalate resin (PET resin), comprising the steps of: irradiating infrared or near-infrared light on the comparative resins of each group, which are group A and group B, which are known to be recycled PET resin, to obtain their respective absorption spectra; extracting spectral data in a predetermined wavenumber range from these absorption spectra; and storing the analytical data obtained by multivariate analysis as a database; and determining whether a resin is recycled PET resin or virgin PET resin as a test resin, which is a resin whose type is unknown as being recycled or virgin PET resin, by extracting spectral data in a predetermined wavenumber range similar to that of the comparative resin to which the test resin is to be compared in the same manner as in the database creation step; comparing the test data with the analytical data group of the database by the multivariate analysis; and using the agreement between the test data and the analytical data group of the database as an index to determine whether the test resin is recycled PET resin or virgin PET resin.

2. For the comparative resins and test resins of each group, a wave number of 4000 cm －１ ~600cm －１ The PET discrimination method according to claim 1, further comprising the steps of: irradiating infrared light in a range of 0.1 to 1.5 nm to obtain the absorption spectrum; extracting spectral data in the specified wavenumber range to obtain the database and the test data; and discriminating whether the test resin is recycled PET resin or virgin PET resin.

3. The predetermined wavenumber range is 1400 cm －１ ~600cm －１ The PET discrimination method according to claim 2, characterized in that the spectral data in these wavenumber ranges is in the range or in a vicinity thereof, and the analysis is performed using the spectral data in these wavenumber ranges.

4. The PET discrimination method according to any one of claims 1 to 3, characterized in that the multivariate analysis is partial least squares discriminant analysis (PLS-DA).

5. The PET discrimination method according to claim 4, characterized in that the PET resin is a fibrous PET fiber.

6. The PET discrimination method according to claim 4, characterized in that the method for determining the absorption spectra of the comparative resin and the test resin is the ATR method (attenuated total reflection method).

7. The PET discrimination method according to claim 6, characterized in that the PET resin is a fibrous PET fiber.

Images