Determination of lactic acid and acetic acid in Baijiu by near-infrared spectroscopy based on HPLC

In order to establish an high performance liquid chromatogram (HPLC) method for determining the contents of lactic acid and acetic acid in Baijiu that was not affected by the volume fraction of ethanol, a near-infrared rapid detection model was established with the reference value combining with kernel partial least squares (KPLS) algorithm. The results showed that using organic polymer packed particle chromatography column for determining the content of lactic acid and acetic acid by HPLC, the relative standard deviations (RSD) of the peak areas of lactic acid and acetic acid were 1.58% and 1.66%, respectively, and the recovery rates were 98.15% and 103.28%, respectively. Near infrared specturm detection model was established, and the optimal preprocessing methods for lactic acid and acetic acid were determined as elimination of constant offset and vector normalization, the optimal additional pretreatment method was Z-score standardization, and competitive adaptive reweighted sampling (CARS) method, and the root mean square error (RMSEP) of the lactic acid and acetic acid test sets for validating the model under optimal conditions was 6.37 mg/100 ml and 5.95 mg/100 ml, respectively, with determination coefficients (R²) of 0.9753 and 0.9591, respectively. The consistency between the predicted values and actual values of external validation samples was good, and the RSD of lactic acid and acetic acid content in repeated prediction samples was 2.28% and 2.22%, respectively. The variance analysis tests indicated that there was no significant difference between the predicted values and the actual values (P>0.05), indicating that the prediction precision of the model was good. In conclusion, the near-infrared spectral model based on the data of HPLC could realize the rapid and accurate determination of lactic acid and acetic acid in Baijiu.