Application of ATR7010 Raman Spectrometer in Edible Oil Adulteration and Origin Identification

2024-01-19

Abstract

This article describes the application of Raman spectroscopy in camellia oil doping and origin identification. A Raman spectrometer was used to detect it, and a baseline correction algorithm was used to establish a model. The results showed that the Raman spectrum intensity has a strong linear relationship with the camellia oil content and can be used to identify the origin of the oil. A model was built based on the quality of camellia oil and Raman light intensity to predict the unknown doping of camellia oil, and good results were obtained, with an accuracy error within 2%.

Application

Edible oil is an efficient source of energy. The human body requires energy support for physiological activities at all times. Each hundred grams of edible oil can produce about 900 kcal of energy, which is more than twice that of carbohydrates and proteins of the same weight. Therefore, It is said that eating oil is an important way for the human body to obtain energy. According to reports, many restaurants, fast food restaurants, and fried chicken shops use oil for deep-frying. Cooking food with waste cooking oil will produce carcinogens, such as benzo(a)pyrene, which is harmful to consumers. Serious risk to health.

Traditional edible oil adulteration detection methods mainly rely on physical and chemical methods, chromatography, gas-mass spectrometry, infrared methods and other detection methods. These detection methods often require tedious pre-processing processes, which are time-consuming, laborious and expensive, and cannot confirm the identity of the origin of oil products, this has caused certain difficulties for some companies and units in adulteration testing and origin identification of edible oils.

Haoquan Jin collected samples of kitchen waste cooking oil and compared them with 5 common edible vegetable oils. The peroxide value, iodine value and nuclear magnetic resonance were measured to reveal the mechanism of the difference in Raman spectra between waste cooking oil and selected edible vegetable oils, combined with chemistry metrological methods to differentiate between waste cooking oil and cooking oil.

Berghian-Grosan C proposed a new method for edible oil identification based on rapid processing of Raman spectra using a machine learning algorithm, which can not only achieve adulteration detection, but also provide a preliminary estimate of the amount of adulteration.

Kwofie F studied the application of Raman spectroscopy and pattern recognition methods to the problem of edible oil type identification. Raman spectral data were collected at a resolution of 2 cm^-1 for 53 edible oil samples of 15 varieties. According to the saturation degree of edible oils and the ratio of polyunsaturated fatty acids to monounsaturated fatty acids, 15 types of edible oils and fats are divided into 5 categories.

Duraipandian S used Raman spectroscopy technology combined with multiple variables to analyze the adulteration problem of extra virgin olive oil and other low-price edible vegetable oils, and established a partial least squares correction model (pls) for binary, ternary and quaternary oil mixtures. It was cross-validated 10-fold to determine its purity, and the purity of the spike after being spiked with one or more cheap oils could be determined. This method is applicable not only to food products, but also to food products obtained from production sites and other food sources.

Shanshan Du et al. used a portable Raman spectrometer to develop a dual-liquid interface plasma array quantitative analyzer for direct classification of edible oil quality. Using principal component analysis to analyze the data collected by the instrument, olive oil can be quickly distinguished. Oxidized oils and adulteration of six types of edible oils: corn oil, rapeseed oil, soybean oil, sunflower oil and linseed oil.

Raman spectroscopy is a scattering spectrum discovered by Indian scientist C. V. Raman in 1928. Its basic principle is that when a beam of monochromatic light with a frequency of υ0 irradiates an unknown sample to be measured, a small part of the photons interacts with the sample to be measured. Energy exchange occurs between the molecules of the substance to be measured, causing the original wave direction of the photons to change, and the frequency also changes from υ0 to υ, and inhibits or excites the molecular vibration of the substance to be measured. This kind of inelastic scattering that can reflect molecular rotation and vibration information is called Raman scattering. By analyzing the Raman shift (that is, the difference between the frequency of incident light and the frequency of scattered light), information on molecular vibration and rotation can be obtained.

The Raman shift has nothing to do with the frequency of the incident light, it is only related to the structure of the scattering molecules themselves. The Raman shift depends on the change of the molecular vibration energy level. Different chemical bonds or molecular groups have their own characteristic molecular vibrations, so the corresponding Raman shift is also characteristic, that is, a characteristic wavelength.

Raman frequency shift is the frequency shift of Raman scattered light relative to the incident light. The frequency shift is determined by the internal structure of the medium, depends on the distribution of molecular vibration energy levels, and has nothing to do with the type, power, and excitation line of the laser light source. Therefore, it is characteristic that different molecules have different vibrational energy levels, and the Raman frequency shift reflects the change of a specific energy level, which is recorded and collected in the form of a spectrum. Based on this, the chemical bonds or groups contained in the molecule can be determined, thereby obtaining the fingerprint information of the molecule for detection purposes.

Raman intensity is related to the nature of chemical bonds and the concentration of substances, so Raman spectroscopy has the ability to qualitatively and quantitatively detect substance components. Moreover, Raman spectroscopy has high molecular selectivity. It can realize single molecule detection and is one of the most sensitive detection technologies. Raman spectroscopy can also be used as a characteristic "molecular fingerprint" to detect inorganic, organic or biological molecules or more complex systems such as biological cells and tissues.

As a widely used fast, non-destructive, and accurate detection and analysis method, Raman spectroscopy technology has gradually become a research hotspot in recent years. It has a wide range of applications, is fast, non-destructive, and non-polluting. Raman spectrum has the characteristics and advantages of being highly characteristic and not limited by the frequency of monochromatic light sources.

Experimental equipment and materials

ATR7010 in-situ Raman spectrometer and quantitative software (Optosky (Xiamen) Co., Ltd.), one ten thousandth electronic balance (LICHEN, model: PA1004), pure camellia oil (Zhishen Lin, Jiuzhou, Hunan), soybean oil (Jinlong fish), 10mm optical path quartz cuvette, disposable plastic tip dropper, dust-free cloth.

Figure 1 Experiment photos.

Experimental part

Use ATR7010 Raman spectrum to detect tea oil, soybean oil and tea oil from different origins. The spectrum information is shown in the figure below:

Figure 2-1 Raman characteristic spectrum of edible oil.

Figure 2-2 Raman characteristic spectra of camellia oil from different origins.

As shown in the figure above: camellia oil and soybean oil have Raman characteristic peaks at wave numbers 1262cm^-1, 1298cm^-1, 1437cm^-1, and 1653cm^-1. 1263cm^-1 and 1653cm^-1 are the characteristic peaks of unsaturated fatty acids, and 1298cm^-1 and 1437cm^-1 are the characteristic peaks of saturated fatty acids. . It can be seen that the doping ratio is different and the intensity of the characteristic peak spectrum is different. Moreover, when camellia oil from different origins is used, the Raman characteristic peak intensities are also different, which can facilitate users and relevant units to identify the origin of edible oils. Using the characteristic peaks of different camellia oils and analyzing the Raman spectral intensities of different concentrations of camellia oil, we can predict the origin and analyze the doping concentration. The analysis results are as follows:

Figure 2-3 The relationship between different concentrations of doped camellia oil and its Raman characteristic peak intensity.

It can be seen that ATR7010 can distinguish and identify the origin of camellia oil. Raman spectrum analysis was performed on the characteristic peaks of edible oil. The intensity of the Raman spectrum has a strong linear relationship with the doping of the oil.

The unit characteristic linear model has large accidental deviations in the analysis, which may bring some inaccurate information when analyzing substances. Collect 20 5% gradient tea oil and soybean oil mixed training samples, use the characteristic peak intensity of tea oil to perform multiple linear regression, and the cross-validation R square reaches 0.99. Using multiple features can significantly increase the robustness of the model compared to using a single feature. The established model is shown in the figure below:

Figure 2-4 Multiple linear regression model.

Soybean oil was doped into tea oil, and tea oil samples were prepared with doping gradients of 9.99%, 19.47%, 30.03%, 39.39%, 50.01%, 59.93%, 69.98%, 79.75%, and 88.68%, and measured Raman light intensity using a Raman spectrometer, and then use the model established by our instrument to verify, and compare the results of the model with the actual results. The results are as follows:

It can be seen that tea oil was adulterated with soybean edible oil, and the Raman spectroscopy method was used for modeling analysis, and the measurement error was within 2%.

Summarize

The use of Raman spectroscopy to model and analyze adulteration in edible oil has laid a solid foundation for Raman's identification of adulteration and origin of edible oil, reflecting the great potential of Raman spectroscopy in regulating the quality of edible oil. And it has huge advantages in solving edible oil adulteration and gutter oil detection.

ATR7010 Raman spectrometer product information

Detector: Ultra-highly sensitive rapid refrigeration cooling type (-10°C) 512-pixel InGaAs detector

Excitation wavelength (optional): 532nm, 785nm, 1064nm

Signal-to-noise ratio: >3000:1

Temperature stability: spectral shift ≤ 1 cm^-1 (10-40 ℃)

Half-peak width: 0.1 nm

Spectral range (cm^-1): 200-2600

Size: about 30cm×22.5cm×13.2 cm; Weight: <10KG

Software functions: data modeling, real-time monitoring and quantitative analysis of substance concentration trends during the reaction process (two-dimensional and three-dimensional graphs)

Application fields: crystallization process, biocatalysis and enzyme catalysis, flow chemistry, polymorph identification, biological process monitoring, chemical synthesis, etc.

Application Cases | Application Of Raman Spectroscopy In Identification Of Cultural Relics

An Exceptional View of Geological Materials with Raman Imaging