rus
Issue #8/2024
S. М. Kuznetsov, V. S. Novikov, D. D. Vasimov, P. K. Laptinskaya, V. V. Kuzmin, M. N. Moskovskiy, E. A. Sagitova
Raman Spectroscopy of Vegetable Oils and Omega 3 Fish Oil Supplements: Quantitative Analysis
Raman Spectroscopy of Vegetable Oils and Omega 3 Fish Oil Supplements: Quantitative Analysis
DOI: 10.22184/1993-7296.FRos.2024.18.8.650.659
This paper discusses the potential of Raman spectroscopy for the characterization of
vegetable oils and dietary supplements based on polyunsaturated fatty acids of the
omega‑3 family (ω‑3 PUFA). The present study demonstrates that Raman spectra can be
employed to determine the iodine value of vegetable oils, form of ω‑3 PUFA, relative total
ω‑3 PUFA content, and ratio of the mass fractions of eicosapentaenoic (EPA) and docosahexaenoic (DHA) acids in the Оmega‑3 dietary supplements.
Furthermore, the paper discusses potential applications of the principal component analysis
(PCA) to Omega‑3 supplement characterization using Raman spectra.
This paper discusses the potential of Raman spectroscopy for the characterization of
vegetable oils and dietary supplements based on polyunsaturated fatty acids of the
omega‑3 family (ω‑3 PUFA). The present study demonstrates that Raman spectra can be
employed to determine the iodine value of vegetable oils, form of ω‑3 PUFA, relative total
ω‑3 PUFA content, and ratio of the mass fractions of eicosapentaenoic (EPA) and docosahexaenoic (DHA) acids in the Оmega‑3 dietary supplements.
Furthermore, the paper discusses potential applications of the principal component analysis
(PCA) to Omega‑3 supplement characterization using Raman spectra.
Теги: dietary supplements iodine value omega‑3 fatty acids principal component analysis raman spectroscopy vegetable oils биологически активные добавки йодное число метод главных компонент омега‑3 кислоты растительные масла спектроскопия комбинационного рассеяния света
Raman Spectroscopy of Vegetable Oils and Omega‑3 Fish Oil Supplements: Quantitative Analysis
S. М. Kuznetsov 1, V. S. Novikov 1, D. D. Vasimov 1, 2, P. K. Laptinskaya 1, V. V. Kuzmin 1, M. N. Moskovskiy 3, E. A. Sagitova 1
Prokhorov General Physics Institute of the RAS, Moscow, Russia.
Moscow Institute of Physics and Technology, Dolgoprudny, Moscow Region, Russia.
Federal Scientific Agronomic and Engineering Center VIM, Moscow, Russia.
This paper discusses the potential of Raman spectroscopy for the characterization of vegetable oils and dietary supplements based on polyunsaturated fatty acids of the omega‑3 family (ω‑3 PUFA). The present study demonstrates that Raman spectra can be employed to determine the iodine value of vegetable oils, form of ω‑3 PUFA, relative total ω‑3 PUFA content, and ratio of the mass fractions of eicosapentaenoic (EPA) and docosahexaenoic (DHA) acids in the Оmega‑3 dietary supplements. Furthermore, the paper discusses potential applications of the principal component analysis (PCA) to Omega‑3 supplement characterization using Raman spectra.
Keywords: dietary supplements, omega‑3 fatty acids, vegetable oils, iodine value, Raman spectroscopy, principal component analysis.
Article received: 14.11.2024
Article accepted: 28.11.2024
Introduction
Today, many industries (medical, pharmaceutical, and agricultural) require the implementation of non-destructive quality control methods for both raw materials and final products. The development of such methods is of a great importance for mass-market products, which include vegetable oils and animal fats. Vegetable oils and animal fats are valuable food products that are widely used in the manufacture of pharmaceuticals and cosmetics [1]. They are also a potential renewable raw material for biodiesel production [1,2] and are main part of the composition of nutritional supplements. In Russia, for example, fish oil supplements (omega‑3 supplements), especially based on omega‑3 polyunsaturated fatty acids (ω‑3 PUFA), are among the most widely consumed categories of supplements [3]. Regular intake of such supplements helps one to compensate the traditional deficiency of ω‑3 PUFAs in Russia, especially the deficiency of eicosapentaenoic acid (EPA, C20H30O2) and docosahexaenoic acid (DHA, C22H32O2), as well as helps to maintain the health of the cardiovascular and nervous systems [4].
Vegetable oils and animal fats are mixtures of triglycerides (TG) of saturated and unsaturated fatty acids. The ratio of the number of unsaturated double (C=C) and saturated single (C–C) carbon-carbon bonds in the molecules of such TGs largely determines many properties of vegetable oils (sensitivity to heat, resistance to oxidation, stability, drying rate, etc.), their storage conditions and applications [5]. This ratio is characterised by the iodine value, defined as the mass of iodine (in grams) that can bond to 100 grams of oil in a chemical halogenation reaction.
Knowledge of the iodine value is essential when using oils in the pharmaceutical, cosmetic and paint industries, as well as in the production of biodiesel [1,2]. However, for a number of applications, particularly in the manufacture of omega‑3 supplements, it is necessary to determine the total ω‑3 PUFA content and the content and ratio of EPA and DHA rather than the iodine value.
The ω‑3 PUFAs found in commercial omega‑3 supplements exist in a number of different forms, which can affect the rate at which the body absorbs the ω‑3 PUFAs. The most prevalent forms of the ω‑3 PUFAs in such supplements are TG and ethyl esters (EE) of fatty acids.
The structural formulas of fatty acid TG and EE molecules, as well as EPA and DHA, are shown in Figure 1.
The counterfeiting or production of low-quality vegetable oils and dietary supplements, as well as violations of transport and storage conditions, pose a significant risk to the safety and health of consumers. Therefore, the development of rapid and efficient analytical methods for these products is essential to ensure quality control, particularly at the stage of delivery from the manufacturer to the consumer, as well as for the in-time detection of counterfeit products.
The traditional techniques for analyzing the composition and quality of vegetable oils and animal fats are chemical and physicochemical methods, namely liquid or gas chromatography and titrimetric analysis [6–8]. These techniques require a high level of expertise and the use of potentially hazardous reagents. In addition, they can only be provided in specialized laboratories. For this reason, alternative methods of oil characterization, based, in particular, on vibrational spectroscopy, are now being actively developed [9–12].
Raman spectroscopy is a fast and non-destructive analytical method that requires no sample preparation. The analysis can be performed in a mobile laboratory or “in situ”. For this reason, this technique is a very attractive option for the study of vegetable oils and animal fats. In contrast to traditional techniques, Raman spectroscopy permits analysis through plastic packaging and, in the case of Omega‑3 supplements, through a gelatin capsule [13]. For the Russian Federation, the important factor in the development of this technique for tasks of product quality monitoring is the established production of portable Raman spectrometers [14].
In this paper, we study the potential of Raman spectroscopy to determine the degree of unsaturation (iodine value) in vegetable oils, as well as the content and form of ω‑3 PUFAs in Omega‑3 supplements.
Materials and methods
This study examined samples of 18 commercial vegetable oils and 18 commercial Omega‑3 dietary supplements. A Senterra II confocal Raman microscope (Bruker, USA) equipped with a 20x objective (N.A. 0.40) and a laser with an emission wavelength of 785 nm and an output power of 100 mW was used to record the Raman spectra of oils and dietary supplements. Spectra were recorded at 1.5 cm‑1 spectral resolution in a 180° scattering geometry. For each sample, 40 spectra were recorded. The accumulation time of each spectrum was 5 s. The iodine values of the vegetable oils was determined by titrimetric analysis using the Margoshes method [15]. The iodine value for each sample was measured 3 to 6 times, and the average value was used in the following analysis.
Principal component analysis was performed using the utility “The Principal Component Analysis for Spectroscopy application” for the OriginPro software package (Learning Edition license).
Results and discussion
Figs. 2 and 3 show the Raman spectra of vegetable oils with different iodine values (IV) and the Raman spectra of Omega‑3 supplements with different relative content of ω‑3 PUFAs, respectively. The relative content was taken as the ratio of the mass of all the PUFAs ω‑3 to the total mass of the fish oil in the supplement. Spectra were normalized to the peak intensity of the Raman band belonging to deformation CH2 vibration (δ(CH2)). This vibration is observed in the spectra of vegetable oils and Omega‑3 supplements with ω‑3 PUFA content less than 40% at 1440 cm‑1. For the content of these acids in the Omega‑3 supplements more than 40%, the δ(CH2)) vibration has a wavenumber of 1448 cm‑1. In the 1200–1700 cm‑1 spectral region, the evolution of the Raman spectra of dietary supplements with increasing ω‑3 PUFA content is similar to that observed for vegetable oils with increasing iodine value (Figs. 2 and 3). Namely, when the degree of unsaturation of the oils and the ω‑3 PUFA content increases, the intensities of the Raman bands with wavenumbers of 1265 and 1658 cm‑1 increase. These vibrations are related to the deformation C–H vibrations and stretching C=C vibrations in HC=CH groups, respectively.
Figs. 4 and 5 show the dependences of the ratio (I1658 / Iδ(СH2)) of the peak intensities of the Raman band at 1658 cm‑1 and the band corresponding to the δ(СH2) vibration on the iodine value (for vegetable oils) and on the relative mass content (in % of the total fat mass) of ω‑3 PUFAs in Omega‑3 dietary supplements, respectively. For both cases, it was found that the experimental values of (I1658 / Iδ(СH2)) are well approximated by linear functions (coefficient of linear correlation r > 0.95). Thus, the (I1658 / Iδ(СH2)) ratio can be used to estimate both the iodine values of vegetable oils and the relative mass content of ω‑3 PUFAs in Omega‑3 supplements. Note, in contrast to vegetable oils, we did not observe a direct proportionality between the ratio (I1658 / Iδ(СH2)) and the ω‑3 PUFA content in dietary supplements. This is due to the contribution to the intensity of the band at 1658 cm‑1 of vibrations of other fatty acids present in in dietary, in particularly monounsaturated fatty acids.
It was found that Raman spectra permit to distinguish the forms (EE or TG) in which ω‑3 PUFAs are present in the supplements (Fig. 1). Namely, a wavenumber shift (~9 cm‑1) in the spectra of Omega‑3 supplements was observed for the stretching vibration of the C=O bonds (ν(C=O)) in addition to the wavenumber shift of the δ(CH2) vibration. In the Raman spectra of dietary supplements with low (less than 40%) ω‑3 PUFA content, as well as in the Raman spectra of vegetable oils, the band corresponding to the ν(C=O) vibration has a maximum at 1747 cm‑1. In the Raman spectra of the dietary supplements with a ω‑3 PUFA content of more than 40%, this vibration was observed at 1738 cm‑1. Note that the manufacturer does not always specify the form of ω‑3 PUFA present in supplements. Among all the studied samples, the form of EE was known only for the supplements containing 73% and 90% ω‑3 PUFAs. As it mentioned above, fatty acids in vegetable oils present in TG form. After comparing the wavenumbers of ν(C=O) vibration for these samples, we concluded that the samples of dietary supplements with the ω‑3 PUFA content of 40–90% contain these acids in the EE form, and the samples with the content of ω‑3 PUFA less than 40% – in the TG form. Therefore, traditional methods of analyzing Raman spectra allow us to evaluate two characteristics of Omega‑3 supplements: the form and the relative total content of ω‑3 PUFAs.
In order to extract information about the ratio of EPA and DHA contents, the Raman spectra of Omega‑3 dietary supplements were analyzed using principal component analysis (PCA). It allows analyzing each spectrum as a whole without selecting particular bands. When applying this method, all the data were initially centered by subtracting the mean spectrum S of the training data set from all spectra Si. After that, a set of principal components (PC) was extracted, and scores were calculated as projections of the centered spectra ΔSi = Si – S onto them. In the score space, each spectrum Si corresponds to a point Oi with coordinates that are the projections of ΔSi onto the PCs [16]. Their mutual arrangement was analyzed using Hotelling’s T2 distribution with the establishment of regions into which points corresponding to samples of a certain group are included with a specified probability. The results of the PCA largely depend on the way the training set of spectra was built and on the selection of the analyzed PCs.
Fig. 6 shows the results of PCA applied to the entire set of Raman spectra of the Omega‑3 supplements studied in this work. In this case, spectral range from 540 to 1800 cm‑1 was analyzed. As can be seen in Figure 6, all the studied dietary supplements were separated into two groups. By comparing the wavenumbers of the ν(C=O) vibration and the groups where the spectra of Omega‑3 supplements appeared on the plane of the first and second principal components (PC1 and PC2, Fig. 6), it was found that the separation occurred according to the forms (EE and TG) of the ω‑3 PUFAs. For clarity, Fig. 6 shows only the part of the pink ellipse in which all the experimental data are concentrated. It should be noted that Doppel Herz and NOW supplement spectra appeared in different groups, in spite of the fact that their ω‑3 PUFA content was almost the same (38 and 40%). Thus, this approach did not provide any new information on the characterization of dietary supplements compared to the data obtained from traditional Raman spectra analysis. For this reason, we excluded the spectral region 1200–1800 cm‑1, which is the most sensitive to the form of ω‑3 PUFA, from consideration and applied the PCA to the Raman spectra of dietary supplements with the EE form of ω‑3 PUFA. According to the information provided by the manufacturers, these supplements differed in the ratio of EPA to DHA masses. Fig. 7 shows the results of PCA applied to the Raman spectra of this group of supplements and to spectral range from 540 to 1200 cm‑1. As it can be seen in Fig. 7, on the plane of the third and fourth principal components (PC3 and PC4), it was possible to divide the supplements according to the ratio of the EPA/DHA mass contents. The ratio of the EPA/DHA mass contents was about the same in supplements containing ω‑3 PUFA in the TG form. Performing a similar analysis for these supplements, we found that none of the points in the PC plane was outside the region in which the spectra of supplements with the ratio of the EPA/DHA mass contents from 1.3 to 1.5 appear with a probability of 95%.
Conclusions
In this study, we revealed that Raman spectroscopy is a rapid tool to determine the degree of unsaturation (iodine value) of vegetable oils. Namely, the ratio of the peak intensities (I1658 / Iδ(СH2)) of the Raman band at 1658 cm‑1 and the Raman band assigned to deformation CH2 vibration (δ(CH2)) is proportional to the iodine value in the range of iodine values from 0 to 188 giodine / 100 goil. This ratio can be used to determine this important characteristic of vegetable oils. In the case of Omega‑3 supplements, traditional methods of analyzing Raman spectra, based on the analysis of individual spectral bands, permit the evaluation of two characteristics: the relative content of all ω‑3 PUFAs and their form in the supplement. This relative content was calculated as the ratio of the mass of all the ω‑3 PUFAs to the total mass of the fish oil in the supplement. We recommend using the (I1658 / Iδ(СH2)) ratio, which has a linear dependence on the relative ω‑3 PUFA content, in order to rapidly estimate this property of the Omega‑3 supplements. Raman monitoring the wavenumber of the C=O stretching vibration (ν(C=O)) allows the differentiation of the forms of ω‑3 PUFAs: the triglyceride form is indicated by the ν(C=O) band at 1 747 cm‑1, while the ethyl ester form is indicated by the ν(C=O) band at 1 738 cm‑1.
The application of principal component analysis allows for the potential expansion of the diagnostic capabilities of Raman spectroscopy in relation to omega‑3 supplements. In particular, the application of this method to the analysis of the spectral range of 540–1200 cm‑1 demonstrated that the third and fourth principal components contain the information regarding the ratio of mass fractions of EPA and DHA. However, the form of ω‑3 PUFAs must be identified before such analysis. Consequently, the development of rapid methods of Raman spectroscopy to determine the quality of omega‑3 supplements should be based on the combination of two approaches to the analysis of Raman spectra: the traditional approach based on the analysis of individual spectral bands and the application of principal component analysis.
The results obtained in this study can be employed in the development of non-destructive analytical techniques for the characterization of the composition and quality of vegetable oils and dietary supplements based on Omega‑3 fatty acids.
Funding Sources
This work was supported by the Russian Science Foundation grant No. 24–22–20100, https://rscf.ru/project/24-22-20100/.
AUTHORS
S. M. Kuznetsov, ORCID: 0000-0002-8378-7085
V. S. Novikov, ORCID: 0000-0002-3304-1568
D. D. Vasimov, ORCID: 0009-0002-8105-0124
P. K. Laptinskaya, ORCID: 0000-0003-1100-0244
V. V. Kuzmin, ORCID: 0000-0002-2434-1817
M. N. Moskovsky, ORCID: 0000-0001-5727-8706
E. A. Sagitova, ORCID: 0000-0001-9992-5879; e-mail: lenochek73@mail.ru
CONTRIBUTION OF THE AUTHORS
S. M. Kuznetsov: design and conduct of the experiment, processing of results, discussions; V. S. Novikov: conducting the experiment, discussions, suggestions and comments; D. D. Vasimov: processing of results, discussions; P. K. Laptinskaya: design and conduct of the experiment, discussions; V. V. Kuzmin: processing of results, discussions; M. N. Moscow: organization of work, suggestions and comments; E. A. Sagitova: idea, processing of results, organization of work, discussions, suggestions and comments.
S. М. Kuznetsov 1, V. S. Novikov 1, D. D. Vasimov 1, 2, P. K. Laptinskaya 1, V. V. Kuzmin 1, M. N. Moskovskiy 3, E. A. Sagitova 1
Prokhorov General Physics Institute of the RAS, Moscow, Russia.
Moscow Institute of Physics and Technology, Dolgoprudny, Moscow Region, Russia.
Federal Scientific Agronomic and Engineering Center VIM, Moscow, Russia.
This paper discusses the potential of Raman spectroscopy for the characterization of vegetable oils and dietary supplements based on polyunsaturated fatty acids of the omega‑3 family (ω‑3 PUFA). The present study demonstrates that Raman spectra can be employed to determine the iodine value of vegetable oils, form of ω‑3 PUFA, relative total ω‑3 PUFA content, and ratio of the mass fractions of eicosapentaenoic (EPA) and docosahexaenoic (DHA) acids in the Оmega‑3 dietary supplements. Furthermore, the paper discusses potential applications of the principal component analysis (PCA) to Omega‑3 supplement characterization using Raman spectra.
Keywords: dietary supplements, omega‑3 fatty acids, vegetable oils, iodine value, Raman spectroscopy, principal component analysis.
Article received: 14.11.2024
Article accepted: 28.11.2024
Introduction
Today, many industries (medical, pharmaceutical, and agricultural) require the implementation of non-destructive quality control methods for both raw materials and final products. The development of such methods is of a great importance for mass-market products, which include vegetable oils and animal fats. Vegetable oils and animal fats are valuable food products that are widely used in the manufacture of pharmaceuticals and cosmetics [1]. They are also a potential renewable raw material for biodiesel production [1,2] and are main part of the composition of nutritional supplements. In Russia, for example, fish oil supplements (omega‑3 supplements), especially based on omega‑3 polyunsaturated fatty acids (ω‑3 PUFA), are among the most widely consumed categories of supplements [3]. Regular intake of such supplements helps one to compensate the traditional deficiency of ω‑3 PUFAs in Russia, especially the deficiency of eicosapentaenoic acid (EPA, C20H30O2) and docosahexaenoic acid (DHA, C22H32O2), as well as helps to maintain the health of the cardiovascular and nervous systems [4].
Vegetable oils and animal fats are mixtures of triglycerides (TG) of saturated and unsaturated fatty acids. The ratio of the number of unsaturated double (C=C) and saturated single (C–C) carbon-carbon bonds in the molecules of such TGs largely determines many properties of vegetable oils (sensitivity to heat, resistance to oxidation, stability, drying rate, etc.), their storage conditions and applications [5]. This ratio is characterised by the iodine value, defined as the mass of iodine (in grams) that can bond to 100 grams of oil in a chemical halogenation reaction.
Knowledge of the iodine value is essential when using oils in the pharmaceutical, cosmetic and paint industries, as well as in the production of biodiesel [1,2]. However, for a number of applications, particularly in the manufacture of omega‑3 supplements, it is necessary to determine the total ω‑3 PUFA content and the content and ratio of EPA and DHA rather than the iodine value.
The ω‑3 PUFAs found in commercial omega‑3 supplements exist in a number of different forms, which can affect the rate at which the body absorbs the ω‑3 PUFAs. The most prevalent forms of the ω‑3 PUFAs in such supplements are TG and ethyl esters (EE) of fatty acids.
The structural formulas of fatty acid TG and EE molecules, as well as EPA and DHA, are shown in Figure 1.
The counterfeiting or production of low-quality vegetable oils and dietary supplements, as well as violations of transport and storage conditions, pose a significant risk to the safety and health of consumers. Therefore, the development of rapid and efficient analytical methods for these products is essential to ensure quality control, particularly at the stage of delivery from the manufacturer to the consumer, as well as for the in-time detection of counterfeit products.
The traditional techniques for analyzing the composition and quality of vegetable oils and animal fats are chemical and physicochemical methods, namely liquid or gas chromatography and titrimetric analysis [6–8]. These techniques require a high level of expertise and the use of potentially hazardous reagents. In addition, they can only be provided in specialized laboratories. For this reason, alternative methods of oil characterization, based, in particular, on vibrational spectroscopy, are now being actively developed [9–12].
Raman spectroscopy is a fast and non-destructive analytical method that requires no sample preparation. The analysis can be performed in a mobile laboratory or “in situ”. For this reason, this technique is a very attractive option for the study of vegetable oils and animal fats. In contrast to traditional techniques, Raman spectroscopy permits analysis through plastic packaging and, in the case of Omega‑3 supplements, through a gelatin capsule [13]. For the Russian Federation, the important factor in the development of this technique for tasks of product quality monitoring is the established production of portable Raman spectrometers [14].
In this paper, we study the potential of Raman spectroscopy to determine the degree of unsaturation (iodine value) in vegetable oils, as well as the content and form of ω‑3 PUFAs in Omega‑3 supplements.
Materials and methods
This study examined samples of 18 commercial vegetable oils and 18 commercial Omega‑3 dietary supplements. A Senterra II confocal Raman microscope (Bruker, USA) equipped with a 20x objective (N.A. 0.40) and a laser with an emission wavelength of 785 nm and an output power of 100 mW was used to record the Raman spectra of oils and dietary supplements. Spectra were recorded at 1.5 cm‑1 spectral resolution in a 180° scattering geometry. For each sample, 40 spectra were recorded. The accumulation time of each spectrum was 5 s. The iodine values of the vegetable oils was determined by titrimetric analysis using the Margoshes method [15]. The iodine value for each sample was measured 3 to 6 times, and the average value was used in the following analysis.
Principal component analysis was performed using the utility “The Principal Component Analysis for Spectroscopy application” for the OriginPro software package (Learning Edition license).
Results and discussion
Figs. 2 and 3 show the Raman spectra of vegetable oils with different iodine values (IV) and the Raman spectra of Omega‑3 supplements with different relative content of ω‑3 PUFAs, respectively. The relative content was taken as the ratio of the mass of all the PUFAs ω‑3 to the total mass of the fish oil in the supplement. Spectra were normalized to the peak intensity of the Raman band belonging to deformation CH2 vibration (δ(CH2)). This vibration is observed in the spectra of vegetable oils and Omega‑3 supplements with ω‑3 PUFA content less than 40% at 1440 cm‑1. For the content of these acids in the Omega‑3 supplements more than 40%, the δ(CH2)) vibration has a wavenumber of 1448 cm‑1. In the 1200–1700 cm‑1 spectral region, the evolution of the Raman spectra of dietary supplements with increasing ω‑3 PUFA content is similar to that observed for vegetable oils with increasing iodine value (Figs. 2 and 3). Namely, when the degree of unsaturation of the oils and the ω‑3 PUFA content increases, the intensities of the Raman bands with wavenumbers of 1265 and 1658 cm‑1 increase. These vibrations are related to the deformation C–H vibrations and stretching C=C vibrations in HC=CH groups, respectively.
Figs. 4 and 5 show the dependences of the ratio (I1658 / Iδ(СH2)) of the peak intensities of the Raman band at 1658 cm‑1 and the band corresponding to the δ(СH2) vibration on the iodine value (for vegetable oils) and on the relative mass content (in % of the total fat mass) of ω‑3 PUFAs in Omega‑3 dietary supplements, respectively. For both cases, it was found that the experimental values of (I1658 / Iδ(СH2)) are well approximated by linear functions (coefficient of linear correlation r > 0.95). Thus, the (I1658 / Iδ(СH2)) ratio can be used to estimate both the iodine values of vegetable oils and the relative mass content of ω‑3 PUFAs in Omega‑3 supplements. Note, in contrast to vegetable oils, we did not observe a direct proportionality between the ratio (I1658 / Iδ(СH2)) and the ω‑3 PUFA content in dietary supplements. This is due to the contribution to the intensity of the band at 1658 cm‑1 of vibrations of other fatty acids present in in dietary, in particularly monounsaturated fatty acids.
It was found that Raman spectra permit to distinguish the forms (EE or TG) in which ω‑3 PUFAs are present in the supplements (Fig. 1). Namely, a wavenumber shift (~9 cm‑1) in the spectra of Omega‑3 supplements was observed for the stretching vibration of the C=O bonds (ν(C=O)) in addition to the wavenumber shift of the δ(CH2) vibration. In the Raman spectra of dietary supplements with low (less than 40%) ω‑3 PUFA content, as well as in the Raman spectra of vegetable oils, the band corresponding to the ν(C=O) vibration has a maximum at 1747 cm‑1. In the Raman spectra of the dietary supplements with a ω‑3 PUFA content of more than 40%, this vibration was observed at 1738 cm‑1. Note that the manufacturer does not always specify the form of ω‑3 PUFA present in supplements. Among all the studied samples, the form of EE was known only for the supplements containing 73% and 90% ω‑3 PUFAs. As it mentioned above, fatty acids in vegetable oils present in TG form. After comparing the wavenumbers of ν(C=O) vibration for these samples, we concluded that the samples of dietary supplements with the ω‑3 PUFA content of 40–90% contain these acids in the EE form, and the samples with the content of ω‑3 PUFA less than 40% – in the TG form. Therefore, traditional methods of analyzing Raman spectra allow us to evaluate two characteristics of Omega‑3 supplements: the form and the relative total content of ω‑3 PUFAs.
In order to extract information about the ratio of EPA and DHA contents, the Raman spectra of Omega‑3 dietary supplements were analyzed using principal component analysis (PCA). It allows analyzing each spectrum as a whole without selecting particular bands. When applying this method, all the data were initially centered by subtracting the mean spectrum S of the training data set from all spectra Si. After that, a set of principal components (PC) was extracted, and scores were calculated as projections of the centered spectra ΔSi = Si – S onto them. In the score space, each spectrum Si corresponds to a point Oi with coordinates that are the projections of ΔSi onto the PCs [16]. Their mutual arrangement was analyzed using Hotelling’s T2 distribution with the establishment of regions into which points corresponding to samples of a certain group are included with a specified probability. The results of the PCA largely depend on the way the training set of spectra was built and on the selection of the analyzed PCs.
Fig. 6 shows the results of PCA applied to the entire set of Raman spectra of the Omega‑3 supplements studied in this work. In this case, spectral range from 540 to 1800 cm‑1 was analyzed. As can be seen in Figure 6, all the studied dietary supplements were separated into two groups. By comparing the wavenumbers of the ν(C=O) vibration and the groups where the spectra of Omega‑3 supplements appeared on the plane of the first and second principal components (PC1 and PC2, Fig. 6), it was found that the separation occurred according to the forms (EE and TG) of the ω‑3 PUFAs. For clarity, Fig. 6 shows only the part of the pink ellipse in which all the experimental data are concentrated. It should be noted that Doppel Herz and NOW supplement spectra appeared in different groups, in spite of the fact that their ω‑3 PUFA content was almost the same (38 and 40%). Thus, this approach did not provide any new information on the characterization of dietary supplements compared to the data obtained from traditional Raman spectra analysis. For this reason, we excluded the spectral region 1200–1800 cm‑1, which is the most sensitive to the form of ω‑3 PUFA, from consideration and applied the PCA to the Raman spectra of dietary supplements with the EE form of ω‑3 PUFA. According to the information provided by the manufacturers, these supplements differed in the ratio of EPA to DHA masses. Fig. 7 shows the results of PCA applied to the Raman spectra of this group of supplements and to spectral range from 540 to 1200 cm‑1. As it can be seen in Fig. 7, on the plane of the third and fourth principal components (PC3 and PC4), it was possible to divide the supplements according to the ratio of the EPA/DHA mass contents. The ratio of the EPA/DHA mass contents was about the same in supplements containing ω‑3 PUFA in the TG form. Performing a similar analysis for these supplements, we found that none of the points in the PC plane was outside the region in which the spectra of supplements with the ratio of the EPA/DHA mass contents from 1.3 to 1.5 appear with a probability of 95%.
Conclusions
In this study, we revealed that Raman spectroscopy is a rapid tool to determine the degree of unsaturation (iodine value) of vegetable oils. Namely, the ratio of the peak intensities (I1658 / Iδ(СH2)) of the Raman band at 1658 cm‑1 and the Raman band assigned to deformation CH2 vibration (δ(CH2)) is proportional to the iodine value in the range of iodine values from 0 to 188 giodine / 100 goil. This ratio can be used to determine this important characteristic of vegetable oils. In the case of Omega‑3 supplements, traditional methods of analyzing Raman spectra, based on the analysis of individual spectral bands, permit the evaluation of two characteristics: the relative content of all ω‑3 PUFAs and their form in the supplement. This relative content was calculated as the ratio of the mass of all the ω‑3 PUFAs to the total mass of the fish oil in the supplement. We recommend using the (I1658 / Iδ(СH2)) ratio, which has a linear dependence on the relative ω‑3 PUFA content, in order to rapidly estimate this property of the Omega‑3 supplements. Raman monitoring the wavenumber of the C=O stretching vibration (ν(C=O)) allows the differentiation of the forms of ω‑3 PUFAs: the triglyceride form is indicated by the ν(C=O) band at 1 747 cm‑1, while the ethyl ester form is indicated by the ν(C=O) band at 1 738 cm‑1.
The application of principal component analysis allows for the potential expansion of the diagnostic capabilities of Raman spectroscopy in relation to omega‑3 supplements. In particular, the application of this method to the analysis of the spectral range of 540–1200 cm‑1 demonstrated that the third and fourth principal components contain the information regarding the ratio of mass fractions of EPA and DHA. However, the form of ω‑3 PUFAs must be identified before such analysis. Consequently, the development of rapid methods of Raman spectroscopy to determine the quality of omega‑3 supplements should be based on the combination of two approaches to the analysis of Raman spectra: the traditional approach based on the analysis of individual spectral bands and the application of principal component analysis.
The results obtained in this study can be employed in the development of non-destructive analytical techniques for the characterization of the composition and quality of vegetable oils and dietary supplements based on Omega‑3 fatty acids.
Funding Sources
This work was supported by the Russian Science Foundation grant No. 24–22–20100, https://rscf.ru/project/24-22-20100/.
AUTHORS
S. M. Kuznetsov, ORCID: 0000-0002-8378-7085
V. S. Novikov, ORCID: 0000-0002-3304-1568
D. D. Vasimov, ORCID: 0009-0002-8105-0124
P. K. Laptinskaya, ORCID: 0000-0003-1100-0244
V. V. Kuzmin, ORCID: 0000-0002-2434-1817
M. N. Moskovsky, ORCID: 0000-0001-5727-8706
E. A. Sagitova, ORCID: 0000-0001-9992-5879; e-mail: lenochek73@mail.ru
CONTRIBUTION OF THE AUTHORS
S. M. Kuznetsov: design and conduct of the experiment, processing of results, discussions; V. S. Novikov: conducting the experiment, discussions, suggestions and comments; D. D. Vasimov: processing of results, discussions; P. K. Laptinskaya: design and conduct of the experiment, discussions; V. V. Kuzmin: processing of results, discussions; M. N. Moscow: organization of work, suggestions and comments; E. A. Sagitova: idea, processing of results, organization of work, discussions, suggestions and comments.
Readers feedback
