Rapid assessment of canola spoilage under sub-optimal storage condition using FTIR spectroscopy

Abstract

The storage environment of grains and oilseeds influences their physico-chemical properties that determine shelf-life and nutritional quality. In case of oilseeds, and more specifically canola, analytical chemistry methods are commonly used to determine their quality which is characterized by fatty acid value (FAV) of samples. As wet chemistry methods are time consuming and require the use of chemicals, Fourier transform infrared (FTIR) spectroscopy combined with multivariate data analysis was investigated for rapid assessment of canola quality as affected by sub-optimal storage. Moreover, in order to conduct the analysis on-site outside of a laboratory setting, the feasibility of using a portable instrument was studied. An FTIR spectrum of canola seeds stored at sub-optimal storage condition (35°C and 84% relative humidity) was obtained weekly for a period of five weeks. The quality degradation over this storage period was measured in terms of reduction in germination and FAV content. Principal components analysis (PCA) was applied on FTIR spectral data for dimensionality reduction and the first two principal components could successfully separate canola samples of different qualities (based on their respective storage durations). Quantitative analysis for prediction of FAV using partial least squares (PLS) regression method was done and models were built utilizing the entire spectral data as well by grouping the spectral into three spectral bands. A root mean square error of prediction (RMSEP) of 4.4% and R2=0.96, was achieved with the model built using the entire mid-infrared region. The spectral bands of 1000–1500 cm-1 and 2500–3000 cm-1 were also able to provide comparable results. Various combinations of spectral pre-processing of data were also explored. The results establish that portable FTIR instruments provide an accurate and rapid alternative to chemical analysis for predicting spoilage and determining canola quality.