傅里叶变换红外吸收光谱识别五种植物油的研究

以花生油、大豆油、芝麻油、棉籽油和米糠油为样品,采用傅里叶变换红外光谱仪,采集傅里叶变换红外吸收光谱,对光谱预处理后,提取红外特征信息,以1746cm-1和2855cm-1处的吸收峰面积比值为横坐标,1099cm-1处与1119cm-1处的吸收峰面积比为纵坐标,在Origin6.0上做出二维分布图,对各种油脂进行识别分析。结果显示,大豆油与其它4种油脂之间有明显区分;大豆油、花生油和芝麻油分布效果好,但棉籽油各样品点之间比较分散;能与其它油脂区分开的有以下几种分布花生油明显区别于芝麻油、棉籽油和大豆油;米糠油明显区别于棉籽油和大豆油。分布有交叉的油脂有米糠油与花生油或芝麻油有交叉,棉籽油与芝麻油有交叉。     

气相色谱-质谱技术结合化学计量学对6种植物油进行判别分析

测定了6种不同种类植物油(茶籽油、大豆油、花生油、葵花籽油、玉米油和芝麻油)的脂肪酸组成及含量,旨在探讨利用植物油脂肪酸的指标对不同种类的植物油进行分类和判别的可能性。采用气相色谱-质谱联用技术,对6种不同植物油中脂肪酸的组成和含量进行测定,用SPSS.19.0统计软件进行主成分分析、聚类分析和判别分析。对6种不同植物油脂肪酸进行分析、对比,得出植物油脂的主要成分为C16∶0,C18∶0,C18∶1 cis-9,C18∶2 cis-9,12和C18∶3 cis-9,12,15。这5种主要脂肪酸的总含量在茶籽油、大豆油、花生油、葵花籽油、玉米油和芝麻油中分别为98.455%,97.586%,89.019%,97.378%,98.294%和98.021%。6种植物油的不饱和度(U/S)均大于2.000,其中最小为花生油2.055,最大为茶籽油3.976。进行主成分分析降维得到前3个主成分,因为前3个主成分的特征//值均大于1且累计贡献率达到80.060%,第1主成分的贡献率为35.853%,第2主成分的贡献率为23.847%,第3主成分的贡献率为20.360%。建立了3个典则判别函数,典则判别函数的相关系数均大于0.990,且对于茶籽油、大豆油、花生油、葵花籽油、玉米油和芝麻油的初始分类正确率为100.0%,交叉验证正确率为100.0%。     

应用化学计量学和顶空-气相色谱质谱联用技术对7种植物油进行类别分析

采用静态顶空-气相色谱质谱联用技术对7种食用植物油(菜籽油、大豆油、花生油、棉籽油、玉米油、芝麻油、棕榈油)中易挥发成分进行了采集,应用化学计量学分析方法对各油种间易挥发成分的差异进行了评价。通过主成份分析、偏最小二乘判别分析等方法,建立了分类预测模型,分类模型训练和模型验证准确率均为100%,鉴定出了每种植物油的主要标志化合物,为建立植物油类别分析提供了强有力的手段。实验证明,根据植物油挥发性物质对不同种类的植物油进行分类和判别是可行的。     

应用校正变换矩阵法识别掺伪食用油

根据食用油中脂肪酸、甾醇以及生育酚含量,应用校正转换矩阵法对花生油掺伪进行了定量检测。用所建立的校正模型对棕榈油、菜籽油和棉籽油掺入到花生油所得到的２７个二元和６个四元人工合成样品进行了验证,结果令人满意。对从市场上购得的５种花生油进行了测定,其中一种是由菜籽油假冒的花生油,另一种为掺入花生精油的棕榈油。     

气相色谱结合化学计量学用于6种食用植物油的分类

运用气相色谱法对6类植物油(大豆油、花生油、茶籽油、菜籽油、玉米油、橄榄油)的脂肪酸组成进行分析,构建植物油的指纹图谱,对植物油进行鉴别和分类。本工作采用遗传-偏最小二乘法(GA-PLS)筛选出7个有效特征变量作为输入变量,采用主成分分析法(PCA)和有监督模式识别法(径向基函数神经网络(RNF-ANN),线性判别分析(LDA)和最小二乘-支持向量机(LSSVM))进行建模分析。结果表明,PCA能够较好地区分六类植物油,而在植物油种类判别分析中,LDA的预报结果最佳。本文提出的方法能够准确直观地区分植物油种类,可用于食用植物油的鉴别和掺杂食用植物油的鉴定。     

气相色谱结合化学计量学分析4种食用植物油的指纹图谱

运用气相色谱法对4类植物油(橄榄油、花生油、菜籽油和大豆油)的脂肪酸组成进行分析,并构建了植物油的指纹图谱,对4类植物油进行鉴别和分类。采用连续投影算法(SPA)对变量进行筛选,选出11个特征变量。以特征变量作为输入,使用主成分分析(PCA)和有监督模式识别(径向基函数神经网络(RBFANN)、线性判别分析(LDA)和最小二乘-支持向量机(LS-SVM))进行建模分析。结果表明,11个特征变量能够较好地区分4类植物油,PCA获得了较好的分类,RBF-ANN的预报结果最佳,预报率为92.6%,并且能准确预报二组分混合掺杂油样。该方法能够准确区分植物油种类,可用于食用植物油的鉴别和掺杂食用植物油的鉴定。     

微波辅助衍生GC-MS测定脂肪酸及校正变换矩阵法用于食用植物油鉴别的研究

建立了微波衍生化GC—MS测定食用植物油中的脂肪酸含量．并采用校正变换矩阵法对食用植物油的成分进行测定的方法。对微波衍生化的实验条件进行了优化．利用建立的校正模型对40个花生油、菜籽油、芝麻油、棉籽油、棕榈油的二元、三元、四元、五元人工合成样品进行了计算预测，所有样品的平均预测误差部小于5％。对8个实际样品进行了测定．其中有3个掺伪样品．所测调和食用油的结果与标签标示结果相符性较好。本方法可用于快速、准确地测定食用油中各成分含量或定性、定量鉴别掺伪成分。     

非正常食用油的在线甲基化-气相色谱测定及模式识别分析研究

建立了食用油中脂肪酸组成的在线水解甲基化-气相色谱测定方法,分析了20余种常用食用油与非正常食用油样品。将1μL(3 mg/mL)油脂样品与2μL衍生化试剂四甲基氢氧化铵(TMAH,25%甲醇溶液)加入裂解器,在350℃下,油脂加水分解瞬间衍生化成相应的脂肪酸甲酯。基于气相色谱图上分离鉴定到的10个共有峰的相对强度,建立了食用油的气相色谱指纹图谱。结合化学模式识别即主成分分析和系统聚类分析对合格食用油和非正常食用油样品的色谱图进行了识别分析。结果表明,建立的指纹图谱结合模式识别技术可以较好的区分合格食用油与非正常食用油样品。     

超高效合相色谱-质谱法快速分析食用植物油中常见脂肪酸

建立了超高效合相色谱-质谱(UPC²-MS)快速分析6种食用植物油(玉米油、葵花籽油、大豆油、茶油、菜籽油、花生油)中棕榈酸、硬脂酸、油酸、亚油酸、亚麻酸等5种常见脂肪酸的方法,并比较了这6种食用油中上述5种脂肪酸的含量差异。采用皂化反应对植物油进行前处理,以ACQUITY UPC² BEH 2-EP色谱柱(100 mm×2.1 mm, 1.7 μm)为分析柱,以超临界CO₂-甲醇/乙腈(1:1, v/v)为流动相进行梯度洗脱,流速为0.8 mL/min。在电喷雾负离子模式下进行检测,外标法定量。结果表明:5种脂肪酸标准物质在0.5~100 mg/L范围内呈现良好的线性关系,相关系数为0.9985~0.9998,定量限(S/N≥10)为0.15~0.50 mg/L;在3个添加水平下,样品的加标回收率为89.61%~108.50%;方法重复性的相对标准偏差(RSD)为0.69%~3.01%。该方法简单、快速、分离效果好,无需对脂肪酸样品进行衍生化,已成功地用于玉米油、葵花籽油、橄榄油、茶油、大豆油和花生油等6种食用油中常见脂肪酸含量的测定。     

应用激光拉曼光谱法检查核桃油的掺假

应用激光拉曼光谱法对掺入其它植物油(如大豆油、花生油、玉米油、棕榈油或棉籽油)的核桃油进行检测。从上述6种纯植物油的拉曼光谱图中取6个主峰的相对拉曼光谱强度为变量进行主成分分析,从所得的二维散点图可知核桃油与其它5种油均分处于图的不同区域;对掺杂不同比例(质量分数)其它油品的核桃油,在其散点图中也可见两者分处不同区域,据此可对掺杂其它油品的核桃油作出定性鉴别。利用波数为1 256cm-1处拉曼光谱相对强度与掺杂油品的含量(w%)之间的线性关系及其所建立的线性回归方程,对掺杂的样品进行了定量分析。在纯核桃油中分别掺入质量分数从8%~95%的棕榈油制备了10份试样,按上述方法进行定量分析,结果表明预测值与已知值的相对误差在1.01%~2.35%之间。     

Effect of ingredients on the mass loss,pasting properties and thermal profile of semi-sweet biscuit dough

The drive to utilise different lipids, both for health benefits and for commercial reasons, in bakery goods has been extensive. However, the roles of the lipid plays in many products, let alone the influence of the level of saturation, are uncertain. The objective of work carried out is to understand how the typical ingredients in biscuit would impact on the thermal profile of semi-sweet biscuit dough. Three different techniques have been used namely gravimetric analysis (TGA), rapid visco analyser and differential scanning calorimetry (DSC). Wheat flour, sugar and fat/oil were the main ingredients used to produce basic dough of semi-sweet biscuit for this study. Semi-sweet biscuit dough formulations with varying types of oils namely palm oil, palm olein, palm stearine, sunflower oil and butterfat were developed. The final mass (i.e. the total amount of moisture lost) for the samples showed significant differences between the doughs; with the control dough, dough contained palm stearine and butter falling into one group and the butter, palm oil, palm olein and sunflower oil forming the second group that showed less mass loss. Doughs containing low levels of saturated fatty acids (palm olein, palm oil and sunflower oil) showed significant difference on the drying properties of samples compared to doughs containing high saturated fatty acids (palm stearine and butter) as revealed by TGA. Pasting properties result showed that oil with different saturation influenced peak viscosity obtained. The DSC results showed that sugar and oil increased the gelatinisation peak temperature up to 2 and 6 °C, respectively. Oils with low saturated fatty acids have more capability to make contact with starch granules during the mixing processing as compared to oil with a high level of saturated fat. It is suggested that the oil presence in the system was delaying the drying process by coating the wheat flour particles hence slowing the drying process as compared to a sample without oil.     

Characterization of Peanut Oil by Infrared Spectroscopy with an Improved Gaussian Mixture Model

The Gaussian mixture model (GMM) and regression (GMR) are widely used statistical tools in pattern classification and nonlinear regression. In this paper, the suitability score was used for variable selection to improve GMM and GMR. The improved GMM was used to characterize peanut oil adulterated with palm oil using Fourier transform infrared spectroscopy. The improved GMR was applied to determine the concentration of palm oil contaminant present. As comparison, GMM and GMR with principal component analysis for feature extraction, support vector machine, back-propagation artificial neural network, nearest centroid classification, and partial least-squares analysis were also used to classify and quantify peanut oil. It was demonstrated that the method is a new classification and regression strategy for the detection of adulterated edible oil.     

Analysis of free fatty acids by a micro-liquid chromatograph directly coupled with a mass spectrometer through a vacuum nebulizing interface

A liquid chromatograph directly coupled with a quadrupole mass spectrometer through a vacuum nebulizing interface was applied to the analysis of various free fatty acids. Chemical ionization mass spectra of the C₇? C₂₂ free fatty acids were first examined using either methanol or benzene as the reagents. Then the practical compositional analysis of the fatty acids were performed with various biological samples such as bean oil, rape oil, palm oil and milk fat where most of the fatty acids are included as their triglycerides.     

Analysis of Peanut Oil Adulterated with Other Edible Oils by Spectrophotometry

Since peanut oil(PO) is more expensive than other seed oils, some PO is adulterated with other cheap seed oils, such as soybean oil, palm olein, cottonseed oil, corn oil and rapeseed oil. The conventional method for determining whether PO was adulterated is to detect the freezing point of oils. The proposed method for the determination of adulterants in PO was based on monitoring the change of absorbance when the sample was refrigerated. A special spectrophotometer was developed. A total of 10 kinds of POs from different suppliers were chosen and adulterated with other seed oils at the volume fraction levels ranging from 5% to 30%. A total of 150 samples were analyzed by the proposed method and the results were satisfactory.     

Using liquid chromatography-mass spectrometry based metabolomics to discriminate between cold pressed rice bran oils produced from two different cultivars of Oryza sativa L. ssp. indica in Thailand

A newly developed liquid chromatography-mass spectrometry (LC-MS) method for the analysis of cold pressed rice bran oil (RBO) was established and used to discriminate between RBOs produced from two different cultivars of major Thai fragrant rice species. The cold pressed RBO was prepared using the screw compression method. The LC-MS data were preprocessed with MZmine 2.10 program before evaluating with principal component analysis using SIMCA 13 software. The LC-MS method was able to detect and quantify several kinds of valuable constituents such as fatty acids, vitamin E, and γ-oryzanol. The chromatographic condition was feasible; short time for analysis and simple method were achieved. From score plot and loading plot of principle component analysis (PCA), two rice cultivar samples were clearly separated, and it was revealed that Khao-Hom-Pathum was more suitable than Khao-Hom-Mali for cold pressed RBO production since it contained high total γ-oryzanol and less saturated free fatty acids. As with the fixed price of all the rice brans, this information can be used in order to, if possible, preserve the price of rice brans from different cultivars.     

Characterisation of spray-dried emulsions with mixed fat phases

Fat encapsulation in spray-dried protein-stabilised emulsions is known to depend on the choice of protein, the emulsion droplet size, and the melting point of the fat. However, the fat encapsulation may also depend on the fat crystal habit. Fats may crystallise in three different forms , β′ and β, of which the β-form is thermodynamically stable. The -form is obtained in rapidly cooled fats, and it can then transform into the β′-form during storage, and this crystal form is finally transformed into the β-form. In order to investigate the effect of different fat phases on the spray-dried emulsions, two solid fats were studied: fully hardened rapeseed oil (β-stable) and fully hardened palm oil (β′-stable). The solid fats were used on their own or in mixtures with rapeseed oil, in order to provide fat phases with different properties. The emulsion composition was chosen as to mimic the composition of whole milk, i.e. 40% lactose, 30% sodium caseinate and 30% fat on a dry weight basis. The dried powders were stored under dry conditions at 4 or 37 °C in order to investigate the changes in the fat crystals and surface composition of the powders with time. The surface composition was analysed using electron spectroscopy for chemical analysis. Evaluation of the data showed that surface coverage of fat varied depending on the composition of the fat phase. The ratio of lactose to protein remained constant, which implies that the fat was present as ‘islands’ on a surface composed of lactose and protein. The hardened palm oil crystallised initially in the - or β′-form (depending on the ratio of hardened fat to oil), and during storage, the crystal form gradually changed into the β′-form. In powders containing hardened rapeseed oil only the stable β-form was found, even in fresh samples. The surface coverage of fat was reduced after storage, whereas the ratio of lactose to protein at the surface remained unchanged. The emulsion droplet size in emulsions prepared at a low homogenisation pressure was considerably increased after spray-drying and reconstitution, whilst the emulsion droplet size was well preserved in emulsions prepared at high homogenisation pressure.     

Feasibility study on the use of infrared spectroscopy for the direct authentication of Iberian pig fattening diet

The feasibility of using both middle- and near-infrared spectroscopy for discrimination between subcutaneous fat of Iberian pigs reared on different fattening diets has been evaluated. The sample set was formed by subcutaneous fat of pigs fattened outdoors (extensively) with natural resources (montanera) and pigs fattened on commercial feeds, either with standard feed or with especial formulations with higher content in oleic acid (HO-formulated feed). Linear discriminant analysis was used to classify the samples according to the fattening diet using the scores obtained from principal component analysis of near- and middle-infrared spectra as variables to construct the discriminant functions. The most influential variables were identified using a stepwise procedure. The discriminant potential of each spectral region was investigated. Best results were obtained with the combination of both regions with 91.7% of the standard feed and 100% of montanera and HO-formulated feed samples correctly classified. Chemical explanations are provided based on the correlation of these variables with fatty acid content in the samples.     

Determination of the fatty acid composition of saponified vegetable oils using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry

A method using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOFMS) for the determination of the fatty acid composition of vegetable oils is described and illustrated with the analysis of palm kernel oil, palm oil, olive oil, canola oil, soybean oil, vernonia oil, and castor oil. Solutions of the saponified oils, mixed with the matrix, meso-tetrakis(pentafluorophenyl)porphyrin, provided reproducible MALDI-TOF spectra in which the ions were dominated by sodiated sodium carboxylates [RCOONa + Na]+. Thus, palm kernel oil was found to contain capric acid, lauric acid, myristic acid, palmitic acid, oleic acid, and stearic acid. Palm oil had a fatty acid profile including palmitic, linoleic, oleic, and stearic. The relative percentages of the fatty acids in olive oil were palmitoleic (1.2 +/- 0.5), palmitic (10.9 +/- 0.8), linoleic (0.6 +/- 0.1), linoleic (16.5 +/- 0.8), and oleic (70.5 +/- 1.2). For soybean oil, the relative percentages were: palmitoleic (0.4 +/- 0.4), palmitic (6.0 +/- 1.3), linolenic (14.5 +/- 1.8), linoleic (50.1 +/- 4.0), oleic (26.1 +/- 1.2), and stearic (2.2 +/- 0.7). This method was also applied to the analysis of two commercial soap formulations. The first soap gave a fatty acid profile that included: lauric (19.4% +/- 0.8), myristic (9.6% +/- 0.5), palmitoleic (1.9% +/- 0.3), palmitic (16.3% +/- 0.9), linoleic (5.6% +/- 0.4), oleic (37.1% +/- 0.8), and stearic (10.1% +/- 0.7) and that of the second soap was: lauric (9.3% +/- 0.3), myristic (3.8% +/- 0.5), palmitoleic (3.1% +/- 0.8), palmitic (19.4% +/- 0.8), linoleic (4.9% +/- 0.7), oleic (49.5% +/- 1.1), and stearic (10.0% +/- 0.9). The MALDI-TOFMS method described in this communication is simpler and less time-consuming than the established transesterification method that is coupled with analysis by gas chromatography/mass spectrometry (GC/MS). The new method could be used routinely to determine the qualitative fatty acid composition of vegetable oils, and, when fully validated by comparison with standard analytical methodologies, should provide a relatively fast quantitative measurement of fatty acid mixtures and/or soap formulations that contain saturated and unsaturated hydrocarbon moieties.