Efficient detection of adulteration in peanut seed oil using a smartphone-based colorimetric sensor array system

This study aimed to develop a smartphone-based colorimetric sensor array (CSA) system for the rapid detection of adulteration in peanut seed oil (PSO). Rapeseed oil (RSO) was mixed with PSO in varying concentrations (5 %, 10 %, 15 %, 20 %, 25 %, and 30 % by weight) to simulate adulteration. The CSA system analyzed the odors of these mixtures and produced distinct color differential maps based on their chemical composition. Multivariate statistical methods, including principal component analysis (PCA), hierarchical cluster analysis (HCA), and k-nearest neighbor (kNN) algorithms, were used to classify and differentiate between pure and adulterated samples. PCA scatter plots and HCA dendrograms revealed clear separations between authentic and adulterated samples, while the kNN model achieved high accuracy in distinguishing the levels of adulteration. These findings highlight the potential of the proposed CSA system as a cost-effective and rapid method for identifying adulteration in PSO. Further validation with a larger dataset is recommended to enhance its reliability and applicability.