Lab-made Chemiresitor Array for Discrimination of Coffee at Different Conditions

摘要

In this paper, we report a multi-nanoparticle chemiresistor array for discrimination of three types of commercially available coffees, in both powder form and liquid form (studied under hot-brewed condition and room temperature). Four types of gold nanoparticles (AuNPs) were deposited on the sensor array via direct drop-casting method. Classification accuracy are compared using four different statistical methods, including linear discriminant analysis (LDA), linear support vector machine (SVM), fine k-nearest neighbors (KNN), and random forest (RF). The results demonstrated the improvement in the overall classification accuracy by using multi-nanoparticle sensing elements compared with devices using single AuNP. The study suggests a promising method to monitor the flavor or quality change of coffee products.