Linear ion trap as a physical tool to address an important problem affecting the integrity of wheat produced in India

• Main Author: Ng, Bee Chin
• Other Authors: Chee Wei Lim
• Format: Final Year Project
• Language: English
• Published: 2014
• Subjects: DRNTU::Science
• Online Access: http://hdl.handle.net/10356/60295
• Institution: Nanyang Technological University

Despite deoxynivalenol being the most commonly found mycotoxin in wheat, little information exists about its surveillance in India. In this paper, flow injection mass spectrometry (FI/MS), without the use of chromatographic separation, was proposed to offer sensitive, high throughput analysis at low costs. We leveraged on the high sensitivity of the linear ion trap in FI/MS to generate information-rich, analyte-specific mass spectra. Approximately 2% of 180 mass spectra were employed in ANN modeling to translate principle m/z values into analyte concentration. The model was trained with 90 spectra of matrix-free calibration curves at 6 concentration levels (50, 100, 200, 300, 400, 500 μg/L). It was then cross validated with 90 spectra of matrix-matched calibration curves, from recovery assessment that spanned two days. They were spiked at 3 concentration levels (100, 200, 300 μg/L). A highly satisfactory recovery range of 83-112% and relative standard deviation of 6% for intra-day and inter-day were achieved. Both models yielded root-mean-square-error of <6%, with square of the correlation coefficients (r2) of >0.9990.The multiple reaction monitoring mode was used as a reference method. Successful implementation of this framework would reinforce high food safety integrity in India.