傅里叶变换红外光谱
偏最小二乘回归
分析化学(期刊)
碘值
红外光谱学
傅里叶变换
样品制备
化学
傅里叶变换光谱学
光谱学
数学
色谱法
有机化学
统计
光学
物理
量子力学
数学分析
作者
F.R. van de Voort,Ashraf A. Ismail,Jacqueline Sedman
摘要
Abstract A rapid Fourier transform infrared (FTIR) method was developed to simultaneously determine percent cis and trans content of edible fats and oils. A generalized, industrial sample‐handling platform/accessory was designed for handling both fats and oils and was incorporated into an FTIR spectrometer. The system was calibrated to predict the cis and trans content of edible oils by using pure triglycerides as standards and partial least squares as the chemometric approach. The efficacy of the calibration was assessed by triglyceride standard addition, by mixing of oils with varying cis/trans contents, and by analyzing fats and oils of known iodine value. Each of the approaches verified that the FTIR method measured the cis and trans content in a reproducible (±0.7%) manner, with the measured accuracies being 1.5% for standard addition and 2.5% for the chemically analyzed samples. Comparisons also were made to the conventional American Oil Chemists’ Society (AOCS) method for the determination of trans isomers by IR spectroscopy. The FTIR‐partial least squares approach worked well over a wide range of trans contents, including those between 0 and 15%. The sample‐handling accessory designed for this application is robust, flexible, and easy to use, being particularly suited for quality‐control applications. In addition, the analysis was automated by programming the spectrometer in Visual Basic (Windows), to provide a simple, prompt‐based user interface and to allow an operator to carry out cis/trans analyses without any knowledge of FTIR spectroscopy. A typical analysis requires less than two minutes per sample. The derived calibration is transferable between instruments, eliminating the need for recalibration. The integrated analytical system provides a sound basis for the implementation of FTIR methods in place of a variety of AOCS wet chemical methods when analytical speed, cost, and environmental concerns are issues.
科研通智能强力驱动
Strongly Powered by AbleSci AI