Rapid and non-destructive ethanol quantification in beer using FTIR-ATR spectroscopy and PLSR modeling

This study presents a rapid, non-destructive, and reliable method for the quantification of ethanol in beer using Fourier Transform Infrared (FTIR) spectroscopy combined with Partial Least Squares Regression (PLSR). FTIR-ATR spectra of beer samples were analysed across four mid-infrared spectral regions (960–1700 cm − 1 ) to identify the optimal range for ethanol prediction. Among the tested intervals, the 1200–1490 cm − 1 region demonstrated minimal matrix interference and yielded results that closely matched those from the reference gas chromatography with flame ionization detection (GC-FID) method. Several PLSR models were evaluated using various spectral preprocessing techniques. The optimal model employed a first derivative transformation, Savitzky–Golay smoothing with a second-order polynomial, and a 21-point window. This model achieved a high coefficient of determination (R 2 = 0.978), with root mean square errors of calibration (RMSEC) and prediction (RMSEP) of 0.578% and 0.36% (v/v), respectively. The method also demonstrated high sensitivity, with a limit of detection (LOD) of 0.021% and a limit of quantification (LOQ) of 0.071%. The findings confirm that FTIR-PLSR is a suitable and efficient alternative to conventional chromatographic techniques for routine ethanol monitoring in the brewing industry.