Method for detecting glutamine supplement adulterated with glycine based on tapered fiber-FTIR

Abstract

The present application relates to the field of spectral detection technology, and more particularly to a kind of based on tapered optical fiber-FTIR glutamine supplement adulteration glycine detection method.The present application includes: preparation of chalcogenide glass tapered optical fiber, coupled to Fourier transform infrared spectrometer external light path, build liquid sample detection platform;Preparation of pure glutamine solution, gradient mixed solution of glycine, low concentration adulteration sample and blind sample, sample and tapered optical fiber contact, collect 4000~400cm ‑1 Infrared absorption spectrum in the range, and determine the characteristic spectral marker of glycine adulteration;Spectral data are preprocessed, and qualitative discrimination model and quantitative prediction model are constructed;Blind sample detection is carried out using quantitative model, and the detection limit and quantitative limit of the method are calculated.The advantage is that the strong evanescent field effect of tapered optical fiber is utilized, the detection sensitivity is significantly improved, and the stable model is constructed by combining chemometrics method, so that rapid and accurate detection of glycine adulteration in glutamine is realized.

Description

Technical Field

[0001] This invention relates to the field of spectroscopic detection technology, and in particular to a method for detecting adulterated glycine in glutamine supplements based on tapered optical fiber-FTIR. Background Technology

[0002] Infrared spectroscopy is a technique that analyzes molecules by utilizing the characteristic absorption of light in specific mid-infrared wavelength regions. When the energy of infrared photons matches the vibrational energy transition energy of sample molecules, the molecules absorb infrared light of that wavelength, resulting in a decrease in the intensity of the transmitted light and the formation of an absorption peak. By analyzing the position of the absorption peak, functional groups and chemical bond types in the sample can be identified, thus enabling qualitative analysis. Mid-infrared fiber optic spectroscopy, by changing the transmission medium of mid-infrared light from a free-space optical path to flexible optical fiber, expands the application boundaries of mid-infrared spectroscopy, offering a wider range of applications compared to traditional mid-infrared spectroscopy techniques.

[0003] Glutamine sports supplements can alleviate muscle damage after high-intensity exercise and boost the body's immunity, preventing discomfort caused by decreased immunity due to strenuous exercise. Because glutamine and glycine have similar physical properties but significantly different production costs, unscrupulous merchants often adulterate or counterfeit glutamine with glycine to reap huge profits. Current methods for identifying adulterated glutamine mainly include high-performance liquid chromatography (HPLC) and traditional mid-infrared spectroscopy. In HPLC, glutamine and glycine have simple structures, lack natural fluorescence, and lack strong ultraviolet absorption groups, making direct high-sensitivity detection impossible. Pre-column or post-column derivatization with reagents such as phthalaldehyde and fluorene chloroformate is required. The derivatization process not only increases operational complexity and result uncertainty, but the reagents used are often highly toxic, volatile, and potentially carcinogenic, posing significant safety risks. Furthermore, this method relies on sophisticated instruments, requires professional operation, and involves expensive, bulky equipment with long detection cycles. Traditional mid-infrared spectroscopy often uses the pellet method or ATR method for detection. The pellet method requires cumbersome steps such as drying, grinding, mixing with potassium bromide, and pelleting, which is time-consuming and requires high operational skills. In addition, potassium bromide is hygroscopic, and the moisture absorption peak seriously interferes with the characteristic peaks of amino acids, affecting the accuracy of the spectrum. Although the ATR method does not require sample preparation, it has limited detection depth and low sensitivity. Summary of the Invention

[0004] To address the safety concerns associated with adulterated glycine in glutamine supplements, this invention provides a method for detecting adulterated glycine in glutamine supplements based on tapered fiber-Fourier transform infrared spectroscopy (FTIR).

[0005] The present invention aims to provide a method for detecting adulterated glycine in glutamine supplements based on tapered optical fiber-FTIR, comprising the following steps:

[0006] S1. Prepare a chalcogenide glass tapered optical fiber, couple it to the external optical path of a Fourier transform infrared spectrometer, and construct a liquid sample detection platform;

[0007] S2. Prepare pure glutamine solution, a mixed solution with gradient glycine doping, a low-concentration adulterated sample, and a blind sample. Place the sample in contact with a tapered optical fiber and collect data at 4000~400cm. -1 The infrared absorption spectrum within the range was analyzed, and characteristic spectral markers of glycine adulteration were identified.

[0008] S3. Preprocess the spectral data to construct a qualitative discrimination model and a quantitative prediction model;

[0009] S4. Use a quantitative model to perform blind sample testing and calculate the limit of detection and limit of quantitation of the method.

[0010] Preferably, the characteristic spectral markers for glycine adulteration include 1350~1300 cm⁻¹. -1 The characteristic peak of the -CH2- bending vibration at 1600 cm⁻¹ -1 Surrounding - COO - Characteristic peak of asymmetric stretching vibration, 1513 cm⁻¹ -1 The surrounding amide II band vibrational characteristic peak or 1419 cm⁻¹ -1 Surrounding - COO - Characteristic peaks of symmetrical stretching vibration.

[0011] Preferably, the characteristic spectral markers for glycine adulteration include 1330 cm⁻¹. -1 The characteristic peak of the -CH2- bending vibration at that location.

[0012] Preferably, the preparation process of the chalcogenide glass tapered optical fiber in step S1 is as follows: select chalcogenide glass optical fiber, and use a precision tapering process to prepare a tapered optical fiber with a waist length of 3~10mm and a waist diameter of 20~40μm.

[0013] Preferably, the components of chalcogenide glass optical fibers include arsenic, selenium, and tellurium.

[0014] Preferably, the liquid sample detection platform includes a Fourier transform infrared spectrometer, an external optical path unit, a liquid sample cell, a tapered optical fiber, a mercury cadmium telluride detector, and a computer; the external optical path unit includes a reflector, a parabolic gold mirror, and a zinc selenide objective lens;

[0015] The mid-infrared beam emitted from the Fourier transform infrared spectrometer is transmitted and focused by an external optical path unit before being coupled into a tapered optical fiber. A mirror in the external optical path unit reflects and redirects the horizontal beam emitted from the Fourier transform infrared spectrometer, a parabolic gold mirror collimates and focuses the beam, and a zinc selenide objective precisely couples the focused beam into the input end of the tapered optical fiber. The tapered optical fiber is fixed within the liquid sample cell, and during operation, the waist region of the tapered optical fiber is in full contact with the liquid sample to be tested. The infrared light modulated by the sample is emitted from the output end of the tapered optical fiber, received by a mercury cadmium telluride detector, converted into an electrical signal, and finally transmitted to a computer to complete the acquisition, storage, and analysis of the spectral data.

[0016] Preferably, the preprocessing of spectral data in step S3 specifically includes: performing baseline correction, difference spectrum processing and SG smoothing in sequence to eliminate baseline drift, matrix background interference and noise signals, thereby improving the effectiveness and signal-to-noise ratio of the spectral data.

[0017] Preferably, the qualitative discrimination model in step S3 is a linear discriminant analysis model, a support vector machine model, or a model built based on random forest, Naive Bayes, or a neural network; the quantitative prediction model is a partial least squares regression model.

[0018] Preferably, the qualitative discrimination model in step S3 is a linear discriminant analysis model or a support vector machine model.

[0019] Compared with the prior art, the present invention can achieve the following beneficial effects:

[0020] This invention discloses a method for detecting adulterated glycine in glutamine supplements based on tapered optical fiber-FTIR. By utilizing the strong evanescent field effect of tapered optical fiber, the detection sensitivity is significantly improved. Combined with chemometric methods to construct a stable model, rapid and accurate detection of glycine adulteration in glutamine supplements can be achieved. Compared with traditional detection methods, this invention does not require complex sample processing or the use of toxic reagents. The detection process is efficient, environmentally friendly, and easy to operate. The specific advantages are reflected in the following three aspects: (1) By utilizing the strong evanescent field effect of tapered optical fiber, the interaction between infrared light and the sample is greatly enhanced, making the characteristic absorption signal of glycine more prominent. The method detection limit is as low as 1.37%, and the quantification limit is 4.56%. It can accurately identify low-concentration adulterants. Even trace amounts of adulteration can be accurately captured, effectively avoiding the problems of missed detection and misjudgment. (2) Samples do not require grinding, tableting, or complex derivatization. They can be directly detected simply by dissolving them in ultrapure water. There is no need to use toxic derivatization reagents, which shortens the detection cycle (the time for a single detection is controlled within tens of seconds), reduces the health risks to operators, and reduces detection costs. (3) By combining the qualitative discrimination model of LDA and SVM with the quantitative model of PLSR, the adulteration ratio can be accurately quantified. The model prediction error is less than 3%, the average error of blind sample detection is 2.47%, and the accuracy rate is 100%. It can meet the needs of rapid screening and batch detection in actual production and is extremely practical. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the liquid sample detection platform provided according to an embodiment of the present invention.

[0022] Figure 2 The figure shows the simulation results of the fundamental mode field distribution of tapered optical fibers with different waist diameters according to an embodiment of the present invention.

[0023] Figure 3 This is the mid-infrared absorption spectrum of a pure glutamine standard sample provided according to an embodiment of the present invention.

[0024] Figure 4 This is a mid-infrared absorption spectrum of a mixed solution of glutamine and glycine with different adulteration ratios, represented by a commercially available brand, according to an embodiment of the present invention.

[0025] Figure 5 This is a schematic diagram illustrating the results of classifying adulterated samples and pure products based on a linear discriminant analysis model according to an embodiment of the present invention.

[0026] Figure 6 This is a correlation diagram between the predicted and actual values ​​of glycine adulteration ratio in three commercially available brand supplements, based on a partial least squares regression model provided in an embodiment of the present invention.

[0027] Figure 7This is the prediction result of adulteration concentration in blind samples provided by an embodiment of the present invention.

[0028] Figure 8 This is a graph showing the detection limit and quantitation limit results of the liquid sample detection platform provided according to an embodiment of the present invention.

[0029] Figure label:

[0030] 1. Fourier transform infrared spectrometer;

[0031] 2. Reflector;

[0032] 3. Parabolic gold mirror;

[0033] 4. Zinc selenide objective lens;

[0034] 5. Liquid sample cell;

[0035] 6. Tapered optical fiber;

[0036] 7. Mercury cadmium telluride detector;

[0037] 8. Computer. Detailed Implementation

[0038] In the following description, embodiments of the invention will be described with reference to the accompanying drawings. In the description below, the same modules are denoted by the same reference numerals. Where the same reference numerals are used, their names and functions are also the same. Therefore, their detailed description will not be repeated.

[0039] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not constitute a limitation thereof.

[0040] This invention provides a method for detecting adulterated glycine in glutamine supplements based on tapered fiber optic-FTIR. Addressing the safety concerns associated with adulterated glutamine supplements, this method combines mid-infrared fiber optic spectroscopy with chemometrics to establish a model capable of identifying and quantifying glycine adulteration in glutamine sports supplements. Unlike current identification methods (high-performance liquid chromatography and traditional mid-infrared spectroscopy), this invention provides a faster, non-destructive, and highly sensitive method for identifying commercial sports supplements.

[0041] The tapered optical fiber of this invention serves as the core detection element, resulting in a significantly lower cost per use compared to conventional mid-infrared detection accessories. Furthermore, it eliminates the need for pellet compression, drastically reducing overall detection costs and sample processing time. Compared to high-performance liquid chromatography (HPLC), this invention eliminates the need for an additional derivatization process, ensuring green detection while also shortening detection time. The detection method of this invention can complete a single spectral acquisition within tens of seconds, offering the advantage of rapid response. Simultaneously, it achieves high detection sensitivity through enhanced evanescent field effects, making it particularly suitable for the rapid identification and quantitative analysis of trace samples. Specifically, it includes:

[0042] S1. Prepare a tapered optical fiber and couple it to the external optical path of a Fourier transform infrared spectrometer (FTIR) to construct a liquid sample detection platform;

[0043] See Figure 1 The liquid sample detection platform includes a Fourier transform infrared spectrometer 1, an external optical path unit, a liquid sample cell 5, a tapered optical fiber 6, a mercury cadmium telluride detector 7, and a computer 8; the external optical path unit includes a reflector 2, a parabolic gold mirror 3, and a zinc selenide objective lens 4.

[0044] The mid-infrared beam emitted from the Fourier transform infrared spectrometer 1 is transmitted and focused by an external optical path unit before being coupled into a tapered optical fiber 6. A reflector 2 in the external optical path unit reflects and redirects the horizontal beam emitted from the spectrometer, a parabolic gold mirror 3 collimates and focuses the beam, and a zinc selenide objective 4 precisely couples the focused beam into the input end of the tapered optical fiber 6. The tapered optical fiber 6 is fixed within the liquid sample cell 5, with its waist region in full contact with the liquid sample to be tested. The enhanced evanescent field of the cone region interacts with the sample molecules. The infrared light modulated by the sample is emitted from the output end of the tapered optical fiber 6, received by a mercury cadmium telluride detector 7, converted into an electrical signal, and finally transmitted to a computer 8 to complete the acquisition, storage, and analysis of the spectral data.

[0045] In some embodiments, the Fourier transform infrared spectrometer 1 is selected as the Bruker INVENIO S model;

[0046] In some embodiments, the method for fabricating tapered optical fibers is as follows: Low-loss chalcogenide glass optical fibers (core diameter 200 μm) are precisely tapered to obtain tapered optical fibers with a waist diameter of 30 μm, a waist length of 5 mm, and a transition region length of 2 mm. Characterization by scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS) confirms that the tapered region of the fiber has a uniform transition, a smooth surface, and a uniform distribution of elements. The composition of the chalcogenide glass optical fiber includes arsenic, selenium, and tellurium, specifically As... 36 Se 48 Te 16 The specific process of drawing tapered optical fibers is as follows:

[0047] S11. Securely clamp both ends of a chalcogenide glass optical fiber onto two moving platforms controlled by ultra-precision linear motors, and adjust the optical fiber to keep it straight.

[0048] S12. At the center of the preset tapered area, a precision heating source is used to locally heat the fiber segment; when the fiber material in the heated area reaches a uniformly softened state, two moving platforms driven by linear motors, under the high-precision synchronous control of the motion controller, begin to move at a constant speed in opposite directions along the fiber axis at a preset speed program.

[0049] S13. Under continuous heating and stretching, the softened fiber region gradually becomes thinner and longer, and is drawn into a tapered fiber with a waist diameter of 30μm; after heating is stopped, the fiber cools and solidifies naturally in the air.

[0050] The drawn tapered optical fiber is installed in the liquid sample cell 5, and the optical path is adjusted to accurately couple the infrared beam into the tapered optical fiber, thus completing the construction of the liquid sample detection platform.

[0051] To investigate the influence of the waist diameter parameter of tapered optical fibers on the enhancement effect of mode field and evanescent field, this invention performs modal analysis on tapered optical fibers with different waist diameters, obtaining the mode field intensity distribution of the fibers under different structural parameters. The results are as follows: Figure 2 As shown. Figure 2 The figure shows the simulation results of mode field intensity distribution for tapered optical fibers with different waist diameters according to the present invention. The horizontal axis represents the fiber diameter (unit: μm), and the vertical axis represents the mode field intensity (unit: au). The curves correspond to tapered optical fibers with waist diameters of 50μm, 40μm, 30μm, and 20μm, respectively. Figure 2 It can be seen that the peak mode field intensity of tapered optical fibers first increases and then decreases as the waist diameter decreases: when the waist diameter is 30 μm, the peak mode field intensity reaches its highest value, significantly higher than that of optical fibers with a waist diameter of 20 μm. Simultaneously, the full width at half maximum (FWHM) of the mode field distribution is wider, indicating that the higher-order mode field intensity at the fiber boundary is maximized under this structural parameter, generating a strong evanescent field for sensing, significantly improving the interaction efficiency between the sample and the optical field, and thus enhancing detection sensitivity. When the waist diameter is greater than 30 μm (e.g., 40 μm, 50 μm), the mode field enhancement effect is insufficient, and the evanescent field intensity is low. When the waist diameter is less than 30 μm (e.g., 20 μm), the thinner the fiber, the lower its mechanical strength, making it extremely prone to breakage, which is detrimental to practical sensing applications. This invention visually presents the spatial distribution and attenuation characteristics of the evanescent field at the fiber boundary with different waist diameters through simulation, ultimately determining 30 μm as the optimal waist diameter parameter for tapered optical fibers. Numerical simulation results show that the conical structure can effectively enhance the intensity of higher-order mode fields at the fiber boundary, generate a strong evanescent field, and significantly improve sensing sensitivity. Optimizing structural parameters through simulation can guide the physical drawing process and reduce trial and error costs.

[0052] S2. Establish a sample set and acquire spectra. Prepare pure glutamine solution, a gradient-doped glycine mixed solution, and blind samples. Place the samples in contact with a tapered optical fiber and acquire spectra at 4000~400cm. -1 Analyze the infrared absorption spectrum within the specified range and determine the characteristic peaks; the specific steps are as follows:

[0053] S21. Sample Set Preparation: Purchase three commercially available glutamine supplements and prepare pure solution samples and iso-concentration gradient adulterated solution samples for model establishment, low-concentration adulterated samples for limit detection, and random-concentration adulterated samples for blind sample detection; wherein:

[0054] (1) Pure solution samples: Using ultrapure water as solvent, L-glutamine standard and three commercially available glutamine supplements were prepared with gradient concentrations ranging from 2 to 20 g / L to establish a pure sample spectral baseline and distinguish between matrix background signals and adulterated signals.

[0055] (2) Adulterated solution samples with equal concentration gradient: Three commercially available glutamine supplements were used as the matrix, and glycine was added to prepare a mixed solution with a total concentration of 10 g / L. The glycine mass adulteration ratio was set from 0% to 100%, with a gradient interval of 10%, to construct the correspondence between the adulteration ratio and the spectral response.

[0056] (3) Low-concentration adulterated samples: Using the first glutamine supplement as the matrix, solutions with glycine adulteration ratios of 1% to 10% and gradient intervals of 1% were prepared, specifically for the evaluation of the method detection limit and quantitation limit;

[0057] (4) Random concentration blind sample: Using the first glutamine supplement as the matrix, the glycine adulteration ratio was determined by a random number generator, and several blind test samples were prepared to verify the generalization ability of the model and the actual detection effect; in the specific embodiment, a total of 10 blind test samples were prepared.

[0058] S22. Spectral Acquisition and Characteristic Peak Determination: Take 0.4 mL of the sample to be tested and add it dropwise into the liquid sample cell, ensuring full contact between the sample and the waist region of the tapered optical fiber. Acquire the spectrum on a Fourier transform infrared spectrometer: with air as the background, the acquisition range is 4000~400 cm⁻¹. -1 The resolution is set to 4cm. -1 The number of scans was 32, completing a single spectral acquisition;

[0059] The collected spectra were analyzed, and the results are as follows: Figure 3 As shown in the figure, the pure glutamine solution without glycine adulteration has a concentration of 1350~1300 cm⁻¹. -1 Within the interval, only extremely weak combination bands exist, with no significant characteristic absorption; such as Figure 4As shown, commercial glutamine supplements adulterated with glycine showed a lower concentration at 1330 cm⁻¹. -1 A distinct characteristic absorption peak appears at 1330 cm⁻¹, the specificity of which originates from the -CH₂- bending vibration of the glycine molecule. Furthermore, this position exhibits no interfering signal in the pure glutamine spectrum, demonstrating extremely strong spectral isolation. Therefore, this invention utilizes the 1330 cm⁻¹... -1 The characteristic peak of the -CH2- bending vibration at the location was identified as a unique and reliable spectral marker for glycine adulteration, which can accurately identify exogenous glycine in the glutamine matrix.

[0060] S3. Preprocess the spectral data to construct qualitative discrimination and quantitative prediction models; specifically including:

[0061] S31. Spectral preprocessing: The acquired raw spectral data is preprocessed by performing baseline correction, difference spectrum processing and SG (Savitzky-Golay) smoothing in sequence to eliminate baseline drift, matrix background interference and noise signals, thereby improving the effectiveness and signal-to-noise ratio of the spectral data.

[0062] S32. Constructing a qualitative discriminant model: Using chemometric methods such as linear discriminant analysis (LDA), support vector machine (SVM), and partial least squares regression (PLSR), with a range of 1350–1300 cm⁻¹... -1 Using preprocessed spectral data within the specified range as input, and with "pure" and "adulterated" as category labels, LDA and SVM discriminant models were constructed respectively. All samples were divided into a training set (158 samples) and an independent test set (52 samples). The model validation results are as follows: Figure 5 As shown in the figure, the horizontal axis LDA1 and the vertical axis LDA2 are the first two discriminant projection directions after dimensionality reduction by the LDA algorithm, namely the first discriminant axis and the second discriminant axis. Each point in the figure represents the projection coordinates of a sample in these two discriminant projection directions. The results show that the LDA method achieves 100% discrimination accuracy on both the training and test sets, and can completely distinguish between pure and adulterated samples.

[0063] S33. Construct a quantitative prediction model: using 1350~1300cm -1 Using preprocessed spectral data within the specified range as input and the mass adulteration ratio of glycine as output, PLSR models were established for samples of three different commercially available glutamine supplement matrices. Leave-one-out cross-validation was used to optimize the model parameters. The model validation results are shown below. Figure 6 As shown, Figure 6 Subplots a, b, and c represent the PLSR models corresponding to the three matrices, where RMSEC represents the root mean square error of the correction set, and Rc... 2 Rp represents the coefficient of determination of the calibration set, RMSEP represents the root mean square error of the prediction, and Rp represents the root mean square error of the prediction. 2R represents the coefficient of determination for the prediction set; the coefficient of determination for prediction (R) 2 The values ​​are all greater than 0.99, and the root mean square error of prediction (RMSEP) is less than 3%, indicating that the model has excellent quantitative prediction accuracy and stability.

[0064] The modeling process in this step was completed in The Unscrambler X 10.4 software to ensure the standardization and repeatability of the model construction.

[0065] S4. Blind Sample Detection and Detection Limit Determination: Verify the practical application performance of the detection model and determine the sensitivity and quantitative reliability of the method, providing data support for the practical promotion of this invention; specifically including:

[0066] S41. Blind Sample Detection and Model Generalization Ability Validation: For the 10 blind adulterated samples prepared in step S2 (glycine adulteration ratio ranging from 10% to 90%), the partial least squares regression (PLSR) quantitative prediction model from step S3 was used for detection. The results are as follows: Figure 7 As shown.

[0067] Figure 7 This is a bar chart comparing the actual adulteration rates of 10 blind samples with the model-predicted adulteration rates. The horizontal axis represents the sample number (A-1 to A-10), and the vertical axis represents the glycine adulteration rate (in %). Dark bars represent the actual adulteration rate, and diagonally lined bars represent the predicted adulteration rate. Figure 7 It can be seen that the predicted values ​​of each blind sample are highly consistent with the actual values. The calculated average error of the model prediction is 2.47%, and the maximum error is 4.28%, which fully demonstrates that the PLSR model established in this invention has excellent generalization ability and practical application value, and can realize the accurate quantitative detection of the adulteration ratio of glycine in unknown commercially available glutamine supplement samples.

[0068] S42. Low-concentration adulteration analysis and determination of detection and quantitation limits: For the first series of low-concentration adulterated samples (glycine mass adulteration ratio ranging from 1% to 10%) of the commercially available glutamine supplement matrix prepared in step S2, infrared spectra were collected and analyzed. A 1330 cm⁻¹ spectral density was selected. -1 The characteristic absorption peak of glycine was located at [location missing]. A calibration curve was established with the absorbance intensity of this peak as the ordinate and the glycine adulteration ratio as the abscissa. The results are as follows: Figure 8 As shown.

[0069] Depend on Figure 8 It can be seen that the absorbance is strongly linearly positively correlated with the glycine adulteration ratio, and the coefficient of determination R of the calibration curve is [value missing]. 2The value was 0.98493. Based on the IUPAC (International Union of Pure and Applied Chemistry) standard, and using the slope of the calibration curve (m = 4.09495E-4) and noise level, the limit of detection (LOD) of this method was calculated to be 1.37%, and the limit of quantitation (LOQ) was 4.56%. These results indicate that the detection method of this invention has high sensitivity and can effectively identify low concentrations of glycine adulteration, meeting the screening requirements for trace adulteration in practical applications.

[0070] In some embodiments, in addition to the three chemometric methods LDA / SVM / PLSR, other classification algorithms such as random forest, Naive Bayes, and artificial neural networks can also be used for discrimination; regression algorithms such as support vector machine regression and principal component regression can also be used for concentration prediction.

[0071] In some embodiments, for the optical path setup of the system, the FTIR spectrometer can be replaced by a portable FTIR or infrared quantum cascade laser, etc.; the light in the optical path is coupled into the tapered fiber after two reflections and two focusings, but increasing or decreasing the number of reflections can also be achieved, the difference being that the intensity of the light effectively coupled into the fiber needs to be adjusted.

[0072] In some embodiments, dividing by 1330cm -1 Characteristic peaks serve as key identification peaks; other characteristic peaks in the adulterated spectrum, such as 1600 cm⁻¹, are also important. -1 Surrounding - COO - Characteristic peak of asymmetric stretching vibration, 1513 cm⁻¹ -1 The surrounding amide II band vibrational characteristic peak, 1419 cm⁻¹ -1 Surrounding - COO - Characteristic peaks of symmetrical stretching vibrations can also be used as marker peaks.

[0073] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this invention disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.

[0074] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for detecting adulterated glycine in glutamine supplements based on tapered optical fiber-FTIR, characterized in that: Includes the following steps: S1. Prepare a chalcogenide glass tapered optical fiber, couple it to the external optical path of a Fourier transform infrared spectrometer, and construct a liquid sample detection platform; S2. Prepare pure glutamine solution, a mixed solution with gradient glycine doping, a low-concentration adulterated sample, and a blind sample. Place the sample in contact with a tapered optical fiber and collect data at 4000~400cm. -1 The infrared absorption spectrum within the range was analyzed, and characteristic spectral markers of glycine adulteration were identified. S3. Preprocess the spectral data to construct a qualitative discrimination model and a quantitative prediction model; S4. Use a quantitative model to perform blind sample testing and calculate the limit of detection and limit of quantitation of the method.

2. The method for detecting adulterated glycine in glutamine supplements based on tapered optical fiber-FTIR according to claim 1, characterized in that: The characteristic spectral markers for glycine adulteration include those in the 1350-1300 cm⁻¹ range. -1 The characteristic peak of the -CH2- bending vibration at 1600 cm⁻¹ -1 Surrounding - COO - Characteristic peak of asymmetric stretching vibration, 1513 cm⁻¹ -1 The surrounding amide II band vibrational characteristic peak or 1419 cm⁻¹ -1 Surrounding - COO - Characteristic peaks of symmetrical stretching vibration.

3. The method for detecting adulterated glycine in glutamine supplements based on tapered optical fiber-FTIR according to claim 2, characterized in that: The characteristic spectral markers of the glycine adulteration include 1330 cm⁻¹. -1 The characteristic peak of the -CH2- bending vibration at that location.

4. The method for detecting adulterated glycine in glutamine supplements based on tapered optical fiber-FTIR according to claim 1, characterized in that: The preparation process of the chalcogenide glass tapered optical fiber in step S1 is as follows: select chalcogenide glass optical fiber, and use a precision tapering process to prepare a tapered optical fiber with a waist length of 3~10mm and a waist diameter of 20~40μm.

5. The method for detecting adulterated glycine in glutamine supplements based on tapered optical fiber-FTIR according to claim 4, characterized in that: The components of the chalcogenide glass optical fiber include arsenic, selenium, and tellurium.

6. The method for detecting adulterated glycine in glutamine supplements based on tapered optical fiber-FTIR according to claim 1, characterized in that: The liquid sample detection platform includes a Fourier transform infrared spectrometer, an external optical path unit, a liquid sample cell, a tapered optical fiber, a mercury cadmium telluride detector, and a computer; the external optical path unit includes a reflector, a parabolic gold mirror, and a zinc selenide objective lens; The mid-infrared beam emitted from the Fourier transform infrared spectrometer is transmitted and focused by an external optical path unit before being coupled into a tapered optical fiber. A mirror in the external optical path unit reflects and redirects the horizontal beam emitted from the Fourier transform infrared spectrometer, a parabolic gold mirror collimates and focuses the beam, and a zinc selenide objective precisely couples the focused beam into the input end of the tapered optical fiber. The tapered optical fiber is fixed within the liquid sample cell, and during operation, the waist region of the tapered optical fiber is in full contact with the liquid sample to be tested. The infrared light modulated by the sample is emitted from the output end of the tapered optical fiber, received by a mercury cadmium telluride detector, converted into an electrical signal, and finally transmitted to a computer to complete the acquisition, storage, and analysis of the spectral data.

7. The method for detecting adulterated glycine in glutamine supplements based on tapered optical fiber-FTIR according to claim 1, characterized in that: The preprocessing of spectral data in step S3 specifically includes: performing baseline correction, difference spectrum processing, and SG smoothing in sequence to eliminate baseline drift, matrix background interference, and noise signals, thereby improving the effectiveness and signal-to-noise ratio of the spectral data.

8. The method for detecting adulterated glycine in glutamine supplements based on tapered optical fiber-FTIR according to claim 1, characterized in that: The qualitative discrimination model in step S3 is a linear discriminant analysis model, a support vector machine model, or a model built based on random forest, Naive Bayes, or a neural network; the quantitative prediction model is a partial least squares regression model.

9. The method for detecting adulterated glycine in glutamine supplements based on tapered optical fiber-FTIR according to claim 8, characterized in that: The qualitative discrimination model in step S3 is either a linear discriminant analysis model or a support vector machine model.

Images