Schulz H, Schrader B, Quilitzsch R, Pfeffer S, Krüger H
Federal Center for Breeding Research on Cultivated Plants, Institute for Plant Analysis, Neuer Weg 22/23, D-06484 Quedlinburg, Germany.
J Agric Food Chem. 2003 Apr 23;51(9):2475-81. doi: 10.1021/jf021139r.
The potential of vibrational spectroscopy methods (attenuated total reflectance/Fourier-transform-infrared (ATR/FT-IR), FT-Raman and near infrared (NIR) spectroscopy) for the identification and quantification of valuable as well as carcinogenic substances in different basil chemotypes is described. It is shown that all main volatile components occurring in different basil accessions can be reliably determined in the isolated essential oils or solvent extracts but also in the air-dried herbs. While NIR data can be interpreted only by chemometric methods, IR and Raman spectra present characteristic key bands of the individual volatiles; therefore, in the latter case, a discrimination of basil chemotypes is frequently possible without applying chemometric algorithms. NIR calibrations are successfully established for various terpenoids and phenylpropanoids; on the basis of these data, the content of the two carcinogenic compounds methyleugenol (range: 2-235 microg/100 g) and estragole (range: 34-138 microg/100 g) can be reliably predicted in air-dried basil leaves (R (2) (coefficient of determination) = 0.951; SECV (standard error of cross validation) = 19.1 microg/100 g and R (2) = 0.890; SECV = 12.8 microg/100 g, respectively). The described methods were found to be very useful tools for the efficient selection of special basil single plants, adapted to the new demands set by the legislator and the consumer. Furthermore, they can be applied in industry to very easily control the purifying, blending, and redistilling processes of basil oil.
描述了振动光谱法（衰减全反射/傅里叶变换红外光谱（ATR/FT-IR）、傅里叶变换拉曼光谱和近红外光谱（NIR））用于鉴定和定量不同罗勒化学型中有价值以及致癌物质的潜力。结果表明，不同罗勒种质中出现的所有主要挥发性成分，不仅可以在分离出的精油或溶剂提取物中，而且可以在风干的草药中可靠地测定。虽然近红外数据只能通过化学计量学方法进行解释，但红外光谱和拉曼光谱呈现出各个挥发性成分的特征性关键谱带；因此，在后一种情况下，通常无需应用化学计量学算法就能区分罗勒化学型。成功建立了针对各种萜类化合物和苯丙素类化合物的近红外校准模型；基于这些数据，可以可靠地预测风干罗勒叶中两种致癌化合物甲基丁香酚（范围：2 - 235微克/100克）和草蒿脑（范围：34 - 138微克/100克）的含量（决定系数R²分别为0.951；交叉验证标准误差SECV = 19.1微克/100克和R² = 0.890；SECV = 12.8微克/100克）。所描述的方法被发现是有效选择特殊罗勒单株植物的非常有用的工具，能够适应立法者和消费者提出的新要求。此外，它们可应用于工业中，以非常轻松地控制罗勒油的纯化、调配和再蒸馏过程。
