Premium
Determination of l‐ theanine in tea water using fluorescence‐visualized paper‐based sensors based on CdTe quantum dots/corn carbon dots and nano‐porphyrin with chemometrics
Author(s) -
Chen Hengye,
Wei Liuna,
Guo Xiaoming,
Hai Chengying,
Xu Lu,
Zhang Lei,
Lan Wei,
Zhou Chunsong,
She Yuanbin,
Fu Haiyan
Publication year - 2021
Publication title -
journal of the science of food and agriculture
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.782
H-Index - 142
eISSN - 1097-0010
pISSN - 0022-5142
DOI - 10.1002/jsfa.10882
Subject(s) - theanine , partial least squares regression , chemometrics , fluorescence , chemistry , detection limit , analytical chemistry (journal) , chromatography , nanosensor , biological system , food science , materials science , mathematics , nanotechnology , green tea , statistics , physics , biology , quantum mechanics
Abstract BACKGROUND The quality of tea is influenced by numerous factors, especially l‐ theanine, which is one of the important markers used to evaluate the sweetness and freshness of tea. Sensitive, rapid, and accurate detection of l ‐theanine is therefore useful to identify the grade and quality of tea. RESULTS A high‐sensitivity, paper‐based fluorescent sensor combined with chemometrics was established to detect l‐ theanine in tea water based on CdTe quantum dots / corn carbon dots and nano tetra pyridel‐porphine zinc (ZnTPyP). To verify the reliability of this method, fluorescence spectra and fluorescence‐visualized paper‐based sensors were compared. The fluorescence spectrum method demonstrated a linear range of 1 to 10 000 nmol L −1 and a limit of detection (LOD) of 0.19 nmol L −1 . In the fluorescence‐visualized paper‐based sensors there was a linear range of 10–1000 nmol L −1 , and the LOD was 10 nmol L −1 . Partial least squares discriminant analysis (PLSDA) and partial least squares regression analysis (PLSR) were used successfully to determine l ‐theanine accurately in tea water with this approach. The accuracy of the PLSDA model was 100% both in the training set and the predicting set, and the correlation coefficient between the actual concentration and the predicted concentration was greater than 0.9997 in the PLSR model. CONCLUSION This fluorescence‐visualized paper‐based sensor, combined with chemometrics, could be applied efficiently to the practical analysis of tea water samples, which provides a new idea to ensure the flavor and quality of tea. © 2020 Society of Chemical Industry