UV–Vis spectroscopy for the estimation of variety and chemical parameters of olive oils

Spectroscopic techniques combined with chemometrics for rapid detection of food adulteration: Applications, perspectives, and challenges

Food adulteration is an important threat to food safety and can be difficult to detect. Some analytical methods are complex and difficult to meet the needs of large numbers of samples. In this study, we introduced the application of six spectroscopic techniques (NIR, FTIR, HSI, Raman, UV-Vis, and FS) and chemometric methods in common food adulteration (powdered food, meat, honey, drink, edible oil, and dairy product) over the last three years. We introduced the consequences of food adulteration, the principles, advantages, and limitations of spectroscopic techniques, spectral data preprocessing and key wavelength selection methods, chemometrics methods, dataset division methods, and evaluation methods for models. Moreover, it provided a perspective for the future application of spectroscopic techniques in food adulteration. The results showed that linear chemometric methods were still the main method used by many researchers, which may limit the application potential of spectroscopic techniques. Therefore, deep learning-based chemometrics methods and their interpretability should be further explored in food adulteration. Secondly, data fusion and ensemble models based on multiple spectroscopic techniques and chemometrics can further improve the accuracy of the models. Future research should select appropriate spectroscopic techniques based on food type and spectroscopic principles, and consider portable technical solutions wherever possible to improve the application scenarios of spectroscopic techniques.

Evaluation of phenolics in the analysis of virgin olive oil using near infrared spectroscopy

Olive oil is an indispensable part of the diet in Mediterranean regions, and is appreciated worldwide for its sensory characteristics, combining a fine aroma and pleasant flavor with the high nutritional value of specific chemical compounds. However, making rapid measurements of phenolic compounds is a major challenge for the olive oil sector. The development of a new method based on near infrared (NIR) spectroscopy may be considered an important advance for the sector, as it is rapid, low-cost, non-contaminant and non-destructive. In this study, three different NIR instruments - one FT-NIR benchtop instrument (671.82-2702.70 nm) and two low-cost portable devices (900-1700 nm and 1350-2150 nm) - were used to analyze a collection of virgin olive oil samples from various Mediterranean regions. To predict both the individual and total concentration of phenols in the olive oil, four signal pretreatment methods and modified partial least squares regression analyses were employed to develop the predictive models. The results showed that the benchtop FT-NIR instrument performed better than the other portable instruments when measuring the phenolic compounds in virgin olive oil. The best models of hydroxytyrosol derivatives, tyrosol derivatives, total phenols and "EFSA phenols" showed R2cv values of 0.84, 0.85, 0.88 and 0.89, while the RPDcv values were 2.51, 2.61, 2.93 and 2.95, respectively. Meanwhile, for the portable NIR instruments, the prediction models for hydroxytyrosol derivatives, total phenols and "EFSA phenols" (the sum of the hydroxityrosol and tyrosol derivatives) showed R2cv values ranging between 0.75 and 0.81 and an RPDcv between 2 and 2.5. These results indicate the great potential of NIR, both with benchtop and portable devices, to detect phenolic compounds in virgin olive oil, which can guarantee the quality of virgin olive oil and thus aid progress in the olive oil industry. Moreover, this is the first published work to determine phenolic compounds in virgin olive oil using portable NIR instruments.

Machine learning-enabled high-throughput industry screening of edible oils

Long-term consumption of mixed fraudulent edible oils increases the risk of developing of chronic diseases which has been a threat to the public health globally. The complicated global supply-chain is making the industry malpractices had often gone undetected. In order to restore the confidence of consumers, traceability (and accountability) of every level in the supply chain is vital. In this work, we shown that machine learning (ML) assisted windowed spectroscopy (e.g., visible-band, infra-red band) produces high-throughput, non-destructive, and label-free authentication of edible oils (e.g., olive oils, sunflower oils), offers the feasibility for rapid analysis of large-scale industrial screening. We report achieving high-level of discriminant (AUC > 0.96) in the large-scale (n ≈ 11,500) of adulteration in olive oils. Notably, high clustering fidelity of 'spectral fingerprints' achieved created opportunity for (hypothesis-free) self-sustaining large database compilation which was never possible without machine learning. (137 words).

Thermal Degradation of Vegetable Oils

Vegetable oils provide lipids and nutrition and provide foods with a desirable flavor, color, and crispy texture when used to prepare fried foods. However, the oil quality is degraded at elevated temperatures, and thus must be examined frequently because of the damage to human health. In this study, sunflower, soybean, olive, and canola oils were examined, and their properties were measured periodically at different elevated temperatures. The unsaturated triglyceride in oils reacted with the environmental oxygen or water vapor significantly changes in optical absorbance, viscosity, electrical impedance, and acid value. We used defect kinetics to analyze the evolution of these oil properties at elevated temperatures. The optical absorbance, viscosity, and electrical impedance follow the second-order, first-order, and zeroth-order kinetics, respectively. The rate constants of the above kinetics satisfy the Arrhenius equation. Olive oil has the lowest rate of color center and dynamic viscosity among the four oils, with the smallest pre-exponential factor and the largest activation energy, respectively. The rate constants of acid reaction also satisfy the Arrhenius equation. The activation energies of the polar compound and acid reaction are almost the same, respectively, implying that the rate constant is controlled by a pre-exponential factor if four oils are compared. Olive oil has the largest rate constant of acid reaction among the four oils, with the lowest pre-exponential factor.

Characterization and Comparison of Extra Virgin Olive Oils of Turkish Olive Cultivars

Extra virgin olive oils (EVOOs) obtained from five Turkish olive cultivars widely produced in the Aegean and Marmara regions were investigated based on their total antioxidant capacity (TAC), total phenolic content (TPC), pigment contents, fatty acid (FA) profiles, phenolic compounds (PC), volatile compounds (VC), and sensory properties. The results showed that all properties of EVOO samples were significantly affected by the olive cultivar used. The pigment contents in Ayvalık (9.90 mg·kg-1) and Uslu (9.00 mg·kg-1) oils were higher than the others (p < 0.05). The greatest values for oleic acid (74.13%) and TPC (350.6 mg·kg-1) were observed in Gemlik and Domat oils, respectively (p < 0.05). Edincik oil showed the maximum hydroxytyrosol content (48.022 mg·kg-1) and TAC value (515.36 mg TE·kg-1) (p < 0.05). The Edincik, Domat, and Uslu oils were significantly not different for the total content of C6 compounds derived by lipoxygenase, which are the main volatiles responsible for the typical aroma of EVOOs (p > 0.05). Domat oil also exhibited the highest scores for bitterness and pungency perceptions (p < 0.05). The fruitiness scores of the oil samples (except for Ayvalık oil) were close to each other, even if they were statistically different (p < 0.05). Principal component analysis (PCA) indicated that the Ayvalık oil was separated from the others due to its poor-quality characteristics. As a result, it can be stated that Domat olive oil has better quality than the others.