In this preliminary study, we used the Multivariate Curve Resolution- Alternating Least Squares (MCR-ALS) algorithm to analyze the excitation-emission matrix for different samples of maize flour contaminated with aflatoxin B1 (AFB1) - uncontaminated, low-contaminated, high-contaminated and flour from the local market. We intended to see if there are differences in emission spectral parameters that depend on degree of contamination. The analysis used genuine emission of the fluorophores in the flour, in absence and presence of AFB1, which enables fast screening of the samples, without sample pre-processing. As a result of the analysis, two fluorescence components were derived from the emission spectra for all analyzed samples. The components' positions were the same for the uncontaminated reference sample and the commercial flour sample from the local market, whereas for the samples contaminated with the aflatoxin B1, the emitted peaks' positions were red-shifted. We found that the ratio of the areas of these two components is proportional to the intensity of contamination: 0.071 for uncontaminated sample, 0.090 for the sample from local market, 0.192 for low-contaminated sample and 1.431 for high-contaminated sample. These results indicate that fluorescence EEM coupled with MCR-ALS could be used for rapid and simple estimation of the degree AFB1 contamination in maize flour.
Keywords: Aflatoxin B1; Excitation-emission matrix; MCR-ALS; Maize flour; Steady state fluorescence spectroscopy.