No abstract

No abstract

No abstract

A colorimetric sensor array composed of synthesized metalloporphyrins selected through quantum-chemical calculation was applied to volatile ethanol quantification.

Near infrared (NIR) spectroscopy combined with high-performance liquid chromatography (HPLC) was applied to discriminate the difference of aromatic vinegar produced during different seasons. 120 samples from 4 seasons were collected, different data preprocessing methods such as standard normal variate transformation (SNV), multiplicative scatter correction (MSC), first derivatives (D1) and second derivatives (D2) were used and linear discriminant analysis (LDA), support vector machine (SVM) and K-nearest neighbor (KNN) models were used to identify the season the aromatic vinegar was produced in. The best result was obtained by the SVM approach with SNV preprocessing, with recognition rate of 100% in training set and test set. The contents of lactic acid, malic acid, L-pyroglutamic acid in aromatic vinegar were analyzed quantitatively through NIR spectroscopy and HPLC combined with partial least squares (PLS) model, backward interval partial least squares (bi-PLS) and back propagation artificial neural network (BP-ANN). The detection of nonlinearity indicated that NIR spectra and the content of lactic acid, malic acid and L-pyroglutamic acid were non-linear and the best results were obtained by BP-ANN model: for lactic acid, optimal PCs was 10, the correlation coefficient (R) for prediction was 0.9342 and root mean square error for prediction (RMSEP) was 0.3310; for malic acid, optimal PCs was 8, R was 0.9337, RMSEP was 0.0557; for L-pyroglutamic acid, optimal PCs was 10, R was 0.9229, RMSEP was 0.0062. The results indicate that NIR spectroscopy could be successfully applied as a rapid method not only to identify the season the aromatic vinegar was produced in, but also to determine the contents of lactic acid, malic acid, L-pyroglutamic acid in aromatic vinegar combined with the BP-ANN model simultaneously.