Method for determining frying oil degradation by near-infrared spectroscopy

A rapid near-infrared (NIR) spectroscopic method for measuring degradation products in frying oils, including total polar materials (TPMs) and free fatty acids (FFAs), has been developed. Calibration models were developed using both forward stepwise multiple linear regression (FSMLR) and partial least-squares (PLS) regression techniques and then tested with two independent sets of validation samples. Derivative treatments had limited usefulness, especially in the longer (1100-2500 nm) wavelength region. When using a wavelength region of 700-1100 nm, PLS models gave improved results compared to FSMLR models. The best correlations (r) between the NIR and wet chemical methods for TPM and FFA were 0.983 and 0.943, respectively. In the longer wavelength region (1100-2500 nm), FSMLR models were as good or better than PLS models. The best correlations for TPM and FFA obtained in this region were 0.999 and 0.983, respectively.     

Does frequent replenishment with fresh monoenoic oils permit the frying of potatoes indefinitely?

The frying life of two monoenoic oils, extra virgin olive oil (EVOO) and high oleic acid sunflower oil (HOSO), used to fry potatoes following a domestic frying model, by replenishing the oil in the fryer with fresh oil after each use to maintain the oil volume/fresh potato ratio, was studied by measuring polar content (PC) and triglyceride oligomer (TO) content. A combination of column and high-performance size-exclusion chromatography was used. Changes in the PC and TO content of both oils according to the number of frying uses were adjusted to linear, logarithmic, and power equations. While all three equations reflected the alterations quite accurately (p < 0.001), the logarithmic and power equations defined them best. Frequent replenishment of both frying oils with fresh oil produced a stabilization of PC and TO levels after 20-40 uses. According to the linear adjustment equation, 321 +/- 33 frying operations with EVOO and 290 +/- 40 with HOSO would be needed to raise PC to 25%. To reach the 10% TO level set for discarding oil used in frying, EVOO and HOSO would have to be used 319 +/- 35 and 241 +/- 37 times, respectively. Using the power adjustment, however, the theoretical frying life of both oils would be much higher (at least 4460 frying uses before reaching a 25% PC and 538-1552 uses for the EVOO and 249-1319 uses for the HOSO to reach a 10% TO level). The frying life of EVOO and HOSO would be extended almost indefinitely (at least 170 000 uses) when the logarithmic adjustment was applied. These data suggest that frequent replenishment of used monoenoic oil with fresh oil permits one to fry sets of fresh potatoes a very high number of occasions.     

近红外光谱快速检测食用油必需脂肪酸

为了建立食用油必需脂肪酸快速检测的方法,该研究提出了基于近红外光谱技术检测食用油中α-亚麻酸和亚油酸含量的快速测定方法。对光谱信息分别采用偏最小二乘回归方法(PLS)和最小二乘支持向量机(LS-SVM)建立模型。比较了多种光谱预处理方法对模型预测能力的影响。结果表明对于亚油酸含量的预测,采用Savitzky-Golay平滑法结合多元散射校正(MSC)的光谱预处理所建立的LS-SVM模型最优。预测集的决定系数(R2)、预测均方根误差(RMSEP)和剩余预测偏差(RPD)分别达到了0.989,0.0161和9.4783。对于α-亚麻酸含量的预测,采用Savitzky-Golay平滑法结合标准正态变换(SNV)的光谱预处理所建立的LS-SVM模型最优。α-亚麻酸含量预测结果的R2、RMSEP和RPD为0.972,0.0036和6.0561,据此表明,应用近红外光谱技术能够检测食用油中α-亚麻酸和亚油酸的含量,为快速检测食用油的必需脂肪酸提供了参考。     

Development of two stable isotope dilution assays for the quantitation of acrolein in heat-processed fats

Two stable isotope dilution assays were developed for the quantitation of acrolein in fats and oils using [(13)C(3)]-acrolein as the internal standard. First, a direct GC-MS headspace method, followed by an indirect GC-MS method using derivatization with pentafluorophenyl hydrazine, was established. Analysis of six different types of oils varying in their pattern of fatty acids showed significant differences in the amounts of acrolein formed after heating at various temperatures and for various times. For example, after 24 h at 140 °C, coconut oil contained 6.7 mg/kg, whereas linseed oil was highest with 242.3 mg/kg. A comparison of the results showed that the extent of acrolein formation seemed to be correlated with the amount of linolenic acid in the oils. Although the acrolein concentrations were lowered in all six oils after frying of potato crisps, linseed and rapeseed oil still contained the highest amounts of acrolein after frying. By applying both methods on different thermally treated fats and oils, nearly identical quantitative data were obtained.     

Geographical classification of honey samples by near-infrared spectroscopy: a feasibility study

The potential of near-infrared (NIR) spectroscopy to determine the geographical origin of honey samples was evaluated. In total, 167 unfiltered honey samples (88 Irish, 54 Mexican, and 25 Spanish) and 125 filtered honey samples (25 Irish, 25 Argentinean, 50 Czech, and 25 Hungarian) were collected. Spectra were recorded in transflectance mode. Following preliminary examination by principal component analysis (PCA), modeling methods applied to the spectral data set were partial least-squares (PLS) regression and soft independent modeling of class analogy (SIMCA); various pretreatments were investigated. For unfiltered honey, best SIMCA models gave correct classification rates of 95.5, 94.4, and 96% for the Irish, Mexican, and Spanish samples, respectively; PLS2 discriminant analysis produced a 100% correct classification for each of these honey classes. In the case of filtered honey, best SIMCA models produced correct classification rates of 91.7, 100, 100, and 96% for the Argentinean, Czech, Hungarian, and Irish samples, respectively, using the standard normal variate (SNV) data pretreatment. PLS2 discriminant analysis produced 96, 100, 100, and 100% correct classifications for the Argentinean, Czech, Hungarian, and Irish honey samples, respectively, using a second-derivative data pretreatment. Overall, while both SIMCA and PLS gave encouraging results, better correct classification rates were found using PLS regression.     

Determination of olive oil oxidative status by selected ion flow tube mass spectrometry

The emergence of primary and secondary oxidation products in New Zealand extra virgin olive oil during accelerated thermal oxidation was measured and correlated with the concentrations of 13 headspace volatile compounds measured by selected ion flow tube mass spectrometry (SIFT-MS). SIFT-MS is a mass spectrometric technique that permits qualitative and absolute quantitative measurements to be made from whole air, headspace, or breath samples in real-time down to several parts per billion (ppb). It is well-suited to high-throughput analysis of headspace samples. Propanal, hexanal, and acetone were found at high concentrations in a rancid standard oil, while propanal, acetone, and acetic acid showed marked increases with oxidation time for the oils used in this study. A partial least-squares (PLS) regression model was constructed, which allowed the prediction of peroxide values (PV) for three separate oxidized oils. Sensory rancidity was also measured, although the correlations of headspace volatile compounds with sensory rancidity score were less satisfactory, and too few results were available for the construction of a PLS regression model. A fast (approximately 1 min), reliable method for prediction of olive oil PVs by SIFT-MS was developed.     

Detecting and quantifying sunflower oil adulteration in extra virgin olive oils from the eastern mediterranean by visible and near-infrared spectroscopy

One hundred and thirty-eight oil samples have been analyzed by visible and near-infrared transflectance spectroscopy. These comprised 46 pure extra virgin olive oils and the same oils adulterated with 1% (w/w) and 5% (w/w) sunflower oil. A number of multivariate mathematical approaches were investigated to detect and quantify the sunflower oil adulterant. These included hierarchical cluster analysis, soft independent modeling of class analogy (SIMCA method), and partial least squares regression (PLS). A number of wavelength ranges and data pretreatments were explored. The accuracy of these mathematical models was compared, and the most successful models were identified. Complete classification accuracy was achieved using 1st derivative spectral data in the 400-2498 nm range. Prediction of adulterant content was possible with a standard error equal to 0.8% using 1st derivative data between 1100 and 2498 nm. Spectral features and chemical literature were studied to isolate the structural basis for these models.     

Comparison of the HPLC method and FT-NIR analysis for quantification of glucose, fructose, and sucrose in intact apple fruits

A rapid quantification method was developed and validated for simultaneous and nondestructive quantifying the constituent sugar concentrations of intact apples using Fourier transform near-infrared (FT-NIR) spectroscopy in diffuse reflectance mode. Multiplicative scatter correction (MSC), the second derivative of Savitsky-Golay, and mean centering were used as spectral preprocessing options. Calibration models were established by the partial least squares (PLS) regression analysis, and validation of the method was performed according to the high-performance liquid chromatography (HPLC) chromatographic method. Spectral range and the number of PLS factors were optimized for the lowest root-mean-square error of prediction (RMSEP) and correlation coefficient of determination (r). The best models showed satisfactory predictions as measured by the RMSEP and r values: glucose, 0.201 and 0.950; fructose, 0.298 and 0.968; sucrose, 0.335 and 0.969, respectively. FT-NIR analysis of constituent sugar concentrations in the intact apple form was found to be more flexible and much faster than performed with the HPLC method.     

Synchronous fluorescence spectroscopy of edible vegetable oils. Quantification of tocopherols

The study demonstrates the application of front face and right angle synchronous fluorescence spectroscopy for the characterization of edible oils. The method enables monitoring of tocopherols, pheophytins, and other fluorescent components in edible oils. Principal component analysis of synchronous fluorescence spectra revealed sample clustering according to the type of oil. Partial least-squares regression was utilized to develop calibration models between fluorescence spectra and total tocopherol content determined by HPLC. The regression models showed a good ability to predict tocopherol content. The best fitting results were obtained for 1% v/v diluted oils and for bulk samples using the entire spectrum, yielding the regression coefficient, r, of 0.991, and root mean square error of cross-validation of approximately 8%. The results presented confirm the capabilities of the fluorescence techniques as a tool for the analysis of edible oils.     

Chemical and sensory evaluation of crude oil extracted from herring byproducts from different processing operations

Fish oils extracted from marinated herring (frozen and unfrozen) byproducts and maatjes herring byproducts were evaluated on their chemical and sensory properties. The obtained crude oils had very low content of copper (<0.1 mg/kg oil), and iron values were 0.8, 0.1, and 0.03 mg/kg oil, respectively, for oil from maatjes and frozen and fresh byproducts. For the maatjes oil, a much lower value was found for alpha-tocopherol compared to the other oils. Storage stability results showed that the oils behave differently. Secondary oxidation products were measured for fresh oil, while for the maatjes and frozen byproducts' oil, tertiary oxidation products were detected. Over storage time, the maatjes and frozen byproducts' oils became more intense in odor, correlating positively at the end with sensory attributes of train-oil, acidic, marine and fishy. The best correlation between sensory and chemical analyses was found for FFA and fishy off-odor (r = 0.781).     

Formation and evolution of monoepoxy fatty acids in thermoxidized olive and sunflower oils and quantitation in used frying oils from restaurants and fried-food outlets

The formation and evolution of monoepoxy fatty acids, arising from oleic and linoleic acids, were investigated in olive oil and conventional sunflower oil, representatives of monounsaturated and polyunsaturated oils, respectively, during thermoxidation at 180 degrees C for 5, 10, and 15 h. Six monoepoxy fatty acids, cis-9,10- and trans-9,10-epoxystearate, arising from oleic acid, and cis-9,10-, trans-9,10-, cis-12,13-, and trans-12,13-epoxyoleate, arising from linoleic acid, were analyzed by gas chromatography after oil derivatization to fatty acid methyl esters. Considerable amounts, ranging from 4.29 to 14.24 mg/g of oil in olive oil and from 5.10 to 9.44 mg/g of oil in sunflower oil, were found after the heating periods assayed. Results showed that the monoepoxides quantitated constituted a major group among the oxidized fatty acid monomers formed at high temperature. For similar levels of degradation, higher contents of the monoepoxides were found in olive oil than in sunflower oil. Ten used frying oils from restaurants and fried-food outlets in Spain were analyzed to determine the contents of the monoepoxides in real frying oil samples. Levels ranged from 3.37 to 14.42 mg/g of oil. Results show that, for similar degradation levels, the monoepoxides were more abundant in the monounsaturated oils than in the polyunsaturated oils.     

拉曼光谱结合UVE-SVR算法预测加热食用油反式脂肪酸的含量

为实现拉曼光谱技术对食用油加热后反式脂肪酸(TFAs)含量的快速预测,将3种食用油在190℃下(常用煎炸温度)进行不同时间加热,每个样品采集36条拉曼光谱。首先,采用多项式平滑与标准正态变量变换(SNN)对原始光谱数据进行预处理,以去除背景和噪音的干扰,然后采用无信息变量消除法(UVE)对光谱数据进行特征变量筛选,最后分别基于全谱数据和筛选的特征光谱变量建立定性和定量分析模型,并对试验结果进行对比分析。结果表明,基于筛选后的变量,运用Fisher判别分析(FDA)建立定性判别模型,其判别正确率由40%~50%提升至90%以上,表明筛选后的变量能较好表征样品的特征信息;分别基于筛选变量和全谱数据,运用偏最小二乘回归(PLSR)、BP神经网络(BPNN)和支持向量回归机(SVR)方法,建立不同样品中TFAs含量的数学预测模型。通过对预测结果对比分析,表明UVE结合SVR方法具有良好的检测效果,菜籽油、大豆油、玉米油的测试集R²分别从0.850 4、0.943 5和0.753 4升至0.952 6、0.954 8和0.958 5。因此,利用UVE-SVR方法不仅简化了预测模型,提高了模型的稳定性和精度,也为食用油中TFA的快速检测提供了一种可行的方法。     

运用PLS算法由HJ-1A/1B遥感影像估测区域小麦实际单产

为进一步提高遥感估产精度,显示国产影像在农业估产中的应用效果。该研究以2010-2013年HJ-1A/1B影像为遥感数据,分析了卫星遥感变量与小麦实际单产的定量关系,运用偏最小二乘回归算法构建及验证了以实际单产为目标的多变量遥感估产模型,并制作了小麦实际单产空间等级分布图。研究表明:实际单产与所选用的大多数遥感变量间关系密切,且多数遥感变量两两间具有严重的多重相关关系;实际单产偏最小二乘回归模型的最佳主成分为5,且植被衰减指数、绿色归一化植被指数、调整土壤亮度的植被指数、比值植被指数和归一化植被指数为实际单产遥感估测的敏感变量;建模集和验证集实际单产估测模型的决定系数分别为0.74和0.70,均方根误差分别为754.05和748.20 kg/hm2,相对误差分别为11.5%和8.88%,且估测精度比线性回归算法分别提高20%以上和40%以上,比主成分分析算法分别提高18%以上和30%以上,说明偏最小二乘回归算法模型估测区域实际单产的效果要明显好于线性回归和主成分分析算法,该模型应用结果与小麦实际单产区域分布情况相符合,为提高区域小麦实际单产的遥感估测精度提供了一种途径。     

Detection of hazelnut oil adulteration using FT-IR spectroscopy

Fourier transform infrared spectroscopy (FT-IR) was used to detect the adulteration of hazelnut oil with different types of oils and to detect the adulteration of extra-virgin olive oil with hazelnut oil. Spectra of hazelnut oil, seven other types of oils, extra-virgin olive oil, and the adulterated oils were collected with a FT-IR equipped with a ZnSe-ATR accessory and a MCTA detector. Discriminant analysis and partial least-squares analysis were used to analyze the data. Classification of hazelnut oil, olive oil, and the other types of oils was achieved successfully with FT-IR. The detection level for sunflower oil adulteration of hazelnut oil was 2%, and the correlation coefficient for the PLS model was 0.99. Adulteration of virgin olive oil with hazelnut oil could be detected only at levels of 25% and higher.     

Measurement of conjugated linoleic acid (CLA) in CLA-rich potato chips by ATR-FTIR spectroscopy

A conjugated linoleic acid (CLA)-rich soy oil has been produced by photoisomerization of soy oil linoleic acid. Nutritional studies have shown that CLA possesses health benefits in terms of reducing certain heart disease and diabetes risk factors. Potato chips are snacks that are readily produced in the CLA-rich soy oil containing CLA levels similar to those of the oil used for frying. The objective of this study was to develop an FTIR method to rapidly determine the CLA content of oil in potato chips. Photoirradiated soy oil samples with ～25% total CLA were mixed with control soy oil, and 100 soy oil samples with total CLA levels ranging from 0.89 to 24.4% were made. Potato chips were fried using each of these 300 g CLA rich soy oil mixtures at 175 °C for approximately 3 min. Duplicate GC-FID fatty acid analyses were conducted on oil extracted from each batch of potato chips. The chip samples were ground and then scanned using ATR-FTIR spectroscopy with the aid of a high-pressure clamp, and duplicate spectra of each sample were averaged to obtain an average spectrum. Calibration models were developed using PLS regression analysis. These correlated the CLA isomer concentrations of potato chips obtained by GC-FID fatty acid analysis with their corresponding FTIR spectral features. The calibration models were fully cross validated and tested using samples that were not used in the calibration sample set. Calibrations for total CLA, trans,trans CLA, trans-10,cis-12 CLA, trans-9,cis-11 CLA, cis-10,trans-12 CLA, and cis-9,trans-11 CLA had coefficients of determinations (R2v) between 0.91 and 0.96 and corresponding root-mean-square error of prediction (RMSEP) ranging from 0.005 to 1.44. The ATR-FTIR technique showed potential as a method for the determination of the CLA levels in unknown potato chip samples.     

Quantitation of short-chain glycerol-bound compounds in thermoxidized and used frying oils. A monitoring study during thermoxidation of olive and sunflower oils

Major short-chain glycerol-bound compounds were investigated in olive oil (OO) and conventional sunflower oil (SO) during thermoxidation at 180 degrees C for 5, 10, and 15 h. These compounds included methyl heptanoate (C7:0), methyl octanoate (C8:0), methyl 8-oxo-octanoate (8-oxo-C8:0), methyl 9-oxononanoate (9-oxo-C9:0), dimethyl octanodiate (C8:0 diester), and dimethyl nonanodiate (C9:0 diester), which were analyzed by GC after derivatization of triacylglycerols to fatty acid methyl esters. An acceptable linear correlation (r = 0.967) was found between the total content of these compounds and the total content of polar compounds, suggesting that quantitation of the major short-chain glycerol-bound compounds provides a good indication of the total alteration level of oils heated at frying temperature. Samples with levels of polar compounds around 25% on oil showed total contents within 2-3 mg/g of oil. To determine the content of these compounds in used frying oils, 10 samples from restaurants and fried-food outlets in Spain were analyzed. Results showed total levels between 2.13 and 7.56 mg/g of oil in samples with contents of polar compounds ranging from 18.8 to 55.5% on oil. Samples with levels of polar compounds of approximately 25% showed total contents of the short-chain compounds similar to those found in the thermoxidized oils, that is, within 2-3 mg/g of oil.     

油炸藕片含油量快速预测及微观结构的三维重建

以油炸藕片为研究对象,首先利用高光谱图像技术(hyperspectral imaging technology,HSIT)结合联合区间偏最小二乘(synergy interval partial least squares,si PLS)和净分析物法(net analyze signal,NAS)提取油炸藕片的光谱信息,建立快速预测模型,并计算每个像素点的油含量值,将其表面油分布可视化;然后利用激光共聚焦显微镜(confocal laser scanning microscopy,CLSM)观察油在细胞中的分布,并扫描藕片Z轴方向序列图片进行微观结构的三维重建。结果显示,优选4个子区间,依据NAS法挖掘同油相关的NAS信号NAsk,建模后其预测集相关系数和均方根误差分别为0.819和0.682 mg/g,表明HSIT可快速预测油炸藕片含油量并可视化表面油分布。经CLSM扫描并图像三维重建后观察,油主要分布在细胞表面、边缘、细胞间隙以及破碎细胞内,且细胞的形状、大小均发生了很大的变化,可观察其内部油分布。结果表明,该研究可为油分布提供有效手段,并依据微观结构的变化为减少含油量提供科学的理论依据。     

光谱指数筛选方法与统计回归算法结合的水稻估产模型对比

为了探寻高效的水稻产量估算方法,在获取2019年黑龙江省三江平原别拉洪河流域内水稻产量数据和MOD09A1遥感数据基础上,对比不同指数筛选方法和统计回归算法结合的建模估产效果,以得到其中最佳的产量估算模型。通过相关系数分析（correlation coefficient,r）法、变量投影重要性分析（Variable Importance in Projection,VIP）法和袋外数据重要性（Out-Of-Bag data importance,OOB）分析法分析水稻4个生育期（分蘖期、抽穗期、孕穗期和乳熟期）的不同波段和光谱指数对于水稻产量的敏感性,筛选出特征波段和指数,再结合随机森林（Random Forest,RF）、支持向量机（Support Vector Machine,SVM）和偏最小二乘（Partial Least Squares,PLS）3种统计回归方法,构建了9种水稻产量估算模型：r-RF、r-SVM、r-PLS、VIP-RF、VIP-SVM、VIP-PLS、OOB-RF、OOB-SVM、OOB-PLS。结果表明：同一指数筛选方法对不同模型的契合程度不同,OOB与RF更为契合,VIP和r与PLS更为契合,r与SVM更为契合;在3种建模方法中偏最小二乘模型和支持向量机模型有较好的效果,随机森林模型效果最好,其中OOB-RF模型最优,其模型验证决定系数为0.742,均方根误差为206 kg/hm2。这为水稻产量估算模型研究提供了关于在指数筛选和回归算法耦合之间的参考,具有一定的理论意义。     

Nondestructive prediction of total phenolics, flavonoid contents, and antioxidant capacity of rice grain using near-infrared spectroscopy

Phytochemicals such as phenolics and flavonoids, which are present in rice grains, are associated with reduced risk of developing chronic diseases such as cardiovascular disease, type 2 diabetes, and some cancers. The phenolic and flavonoid compounds in rice grain also contribute to the antioxidant activity. Biofortification of rice grain by conventional breeding is a way to improve nutritional quality so as to combat nutritional deficiency. Since wet chemistry measurement of phenolic and flavonoid contents and antioxidant activity are time-consuming and expensive, a rapid and nondestructive predictive method based on near-infrared spectroscopy (NIRS) would be valuable to measure these nutritional quality parameters. In the present study, calibration models for measurement of phenolic and flavonoid contents and antioxidant capacity were developed using principal component analysis (PCA), partial least-squares regression (PLS), and modified partial least-squares regression (mPLS) methods with the spectra of the dehulled grain (brown rice). The results showed that NIRS could effectively predict the total phenolic contents and antioxidant capacity by PLS and mPLS methods. The standard errors of prediction (SEP) were 47.1 and 45.9 mg gallic acid equivalent (GAE) for phenolic content, and the coefficients of determination ( r (2)) were 0.849 and 0.864 by PLS and mPLS methods, respectively. Both PLS and mPLS methods gave similarly accurate performance for prediction of antioxidant capacity with SEP of 0.28 mM Trolox equivalent antioxidant capacity (TEAC) and r (2) of 0.82. However, the NIRS models were not successful for flavonoid content with the three methods ( r (2) < 0.4). The models reported here are usable for routine screening of a large number of samples in early generation screening in breeding programs.     

Comparison of volatile aldehydes present in the cooking fumes of extra virgin olive, olive, and canola oils

Emissions of low molecular weight aldehydes (LMWAs) from deep-frying of extra virgin olive oil, olive oil, and canola oil (control) were investigated at two temperatures, 180 and 240 degrees C, for 15 and 7 h, respectively. The oil fumes were collected in Tedlar bags and then analyzed by gas chromatography-mass spectrometry. Seven alkanals (C2-C7 and C9), eight 2-alkenals (C3-C10), and 2,4-heptadienal were found in the fumes of all three cooking oils. The generation rates of these aldehydes were found to be dependent on heating temperature, showing significant increases with increases in temperature. The LMWA emissions from both kinds of olive oils were very similar and were lower than those observed from canola oil under similar conditions. These results suggest that frying in any type of olive oil, independent of its commercial category, will effectively decrease the generation of volatile aldehydes in the exhaust. This fact is important because less expensive refined olive oil is usually used for deep-frying operations, whereas extra virgin olive oil is usually used as salad dressing.