Non-targeted Detection of Multiple Frauds in Orange Juice Using Double Water-Soluble Fluorescence Quantum Dots and Chemometrics

The feasibility of a highly sensitive "turn-off" fluorescent probe of double quantum dots (QDs) combined with chemometrics was investigated for untargeted screening of extraneous adulterants in pure orange juice (OJ), including sucrose syrup and artificial fruit powder. Pure and adulterated OJ samples were characterized by their different quenching patterns of the two separate and strong fluorescent peaks generated by the double QDs followed by chemometrics analysis. Class models of pure OJ samples (n = 117) obtained from pressing newly harvested oranges were developed using one-class partial least squares (OCPLS) based on different signal preprocessing methods, including smoothing, taking second-order derivatives (D2) and standard normal variate (SNV) transformation. As a result, D2-OCPLS model could detect at 5.0% (w/w) of sucrose syrup and 2.0% (w/w) of artificial fruit powder in pure OJ with a sensitivity (the rate of true positives) of 97.8% and specificity (rate of true negatives) of 77.0%. In conclusion, the proposed fluorescence probe with double QDs has been demonstrated to have potential for applications in rapid and sensitive screening of adulterants in OJ, which also implies promising applications to untargeted analysis of other water-soluble food samples.