|本期目录/Table of Contents|

基于傅里叶变换红外光谱技术检测奶粉中香兰素的新方法(PDF)

《纳米技术与精密工程》[ISSN:1672-6030/CN:12-1351/O3]

期数:
2017年6期
页码:
438-443
栏目:
出版日期:
2017-11-15

文章信息/Info

Title:
Development of a Novel Method for Detecting Vanillin in Milk Powder Based on FT-IR Spectroscopy
文章编号:
1672-6030(2017)06-0438-06
作者:
陈达 邹建 谭棕 刘俊鑫 李奇峰
天津大学精密仪器与光电子工程学院, 天津 300072
Author(s):
Chen Da Zou Jian Tan Zong Liu Junxin Li Qifeng
School of Precision Instrument and Opto-Electronics Engineering, Tianjin University, Tianjin 300072, China
关键词:
掺杂奶粉检测 香兰素 傅里叶变换红外光谱 离散小波变换 偏最小二乘法
Keywords:
detection of adulterated milk powder vanillin FT-IR spectroscopy discrete wavelet transform partial least squares
分类号:
O657.33;TS207
DOI:
DOI 10.13494/j.npe.20160100
文献标识码:
A
摘要:
违规使用香兰素是影响婴幼儿奶粉食用安全的重要因素之一,对其进行快速检测具有重要的意义.本文采用长光程傅里叶变换红外光谱(FTIR)以顶空采样的方式,对掺杂香兰素的奶粉体系中的挥发性气体进行高灵敏检测,巧妙规避了奶粉复杂基质对香兰素分析的干扰,并显著提升了香兰素检测的灵敏度.为了进一步提升定量分析方法的检测灵敏度,本文发展了多尺度建模方法,该方法有机结合了离散小波变换(DWT)和偏最小二乘法(PLS),充分利用气体红外光谱中的时/频多尺度信息,从复杂、变动的奶粉红外光谱中准确提取微弱的香兰素吸收信息.结果表明,DWTPLS算法可显著提升模型的预测精度和可靠性,进而推动长光程红外光谱检测技术在奶粉安全检测中的应用.
Abstract:
The illegal use of food additives represents one of the most important factors that influence the safety of milk powder, thus the rapid detection of illegal additives is of great significance. This paper develops a strategy with the combination of the long path Fourier transform infrared (FT-IR) spectroscopy and headspace sampling to detect volatile gas of milk powder doped with vanillin, thus skillfully avoiding the interference of complex solid matrix, which would significantly improve the detection sensitivity of vanillin. To further improve the sensitivity of quantitative analysis process, a multi-scale modeling method was developed. The method combines discrete wavelet transform with partial least squares method (DWT-PLS), making full use of the gas infrared spectrum information in multiscale time/frequency domains to extract weak absorption information of vanillin in the presence of complex milk powder matrix. Results reveal that the DWT-PLS algorithm can significantly improve the prediction accuracy and reliability, thus promoting the application of long path infrared spectroscopy technology in the milk powder inspection.

参考文献/References

备注/Memo

备注/Memo:
收稿日期: 2017-08-24. 基金项目: 国家自然科学基金资助项目(21305101). 作者简介: 陈达(1978—),男,博士,研究员. 通讯作者: 陈达,dachen@tju.edu.cn.
更新日期/Last Update: 2017-12-21