This study developed a novel colorimetric sensor array (CSA) based on mesoporous silica nanospheres (MSNs) for the quantitative detection of aflatoxin B1 (AFB1) in wheat. First, 12 kinds of nano-composite color-sensitive materials (Nano-CSMs) were prepared based on MSNs combined with metal porphyrin dyes, and their color-sensitive materials (CSMs) were modified. The volatile gas information of wheat samples with different degrees of mold was collected using two different sensor arrays, and the characteristic maps were visualized. Next, the genetic algorithm (GA) and particle swarm optimization (PSO) were used to screen the optimal characteristic components, and a support vector regression (SVR) model was established to achieve rapid quantitative detection of AFB1. The optimization results show that the SVR model based on Nano-CSA performed best, with a correlation coefficient (Rp) of 0.9754. The results show that the colorimetric sensor array modified by MSNSs has higher sensitivity and efficiency in the detection of AFB1 in wheat.
Read full abstract