Aroma visualization: A cutting-edge sensor for evaluating the roasting quality of large-leaf yellow tea

Roasting is critical for forming the unique roasted flavor of large-leaf yellow tea (LYT). However, rapid and scientific methods for monitoring the quality of roasting have not yet been developed. In this study, novel colorimetric sensors based on nano-modified and porous silica nanosphere (PSN)/metal organic framework (MOF) porous materials modified porphyrin (TPP) dyes were proposed for monitoring the roasting quality of LYT. First, aroma compounds of LYT with different roasting quality was identified by headspace solid phase microextraction-gas chromatography-mass spectrometry and four TPPs were screened according to their response to LYT aroma. Scanning electron microscope (SEM), transmission electron microscope (TEM) and energy dispersive spectrometer (EDS) were used to characterize nanoporphyrin (N-TPP) and PSN/MOF. The RGB and hyperspectral response features of different sensing arrays were then extracted, and the performance of extreme learning machine (ELM), least square support vector machine (LSSVM) and convolutional neural network (CNN) algorithms for processing sensing array data were compared. The CNN shows the highest discriminant rate. Both the PSN/MOF@N-TPP-based CNN models achieved a discriminant rate of 100%, exceeding the performance of the TPP-based CNN model (90%). The proposed method can effectively improve the accuracy of monitoring the roasting quality of LYT. • A novel aroma sensing method for discriminating LYT roasting quality is proposed. • GC-MS results show the effect of LYT roasting quality on aroma. • Nanosizing and PSN/MOF-modification are used to improve the sensitivity of TPP. • Deep learning algorithm CNN shows advantages in CSA data processing. • The optimal CSA discriminates the quality of LYT roasting with 100% accuracy.