A rapid automated method for wine analysis based upon sequential injection (SI)-FTIR spectrometry

Abstract

A new process control methodology for the simultaneous determination of sugars, alcohols and organic acids in wine based on multivariate evaluation of mid-IR transmission spectra of wine samples is presented. In addition to ethanol several lower level wine components (glucose, fructose, glycerol, citric-, tartaric-, malic-, lactic- and acetic acid) were determined. To establish a multivariate calibration model a set of 72 calibration solutions was prepared and measured, using a novel, fully automated sequential injection (SI) system with Fourier transform infrared (FTIR) detection. The resulting spectra were evaluated using a partial least square (PLS) model. The developed PLS model was then applied to the analysis of real wine samples containing 79–91 g L–1 ethanol, 5.9–8.1 g L–1 glycerol, 0.4–6.9 g L–1 glucose, 1.5–7.5 g L–1 fructose, 0.3–1.6 g L–1 citric acid, 1.0–1.7 g L–1 tartaric acid, 0.02–3.2 g L–1 malic acid, 0.4–2.8 g L–1 lactic acid and 0.15–0.60 g L–1 acetic acid, yielding results which were in good agreement with those obtained by an external reference method (HPLC-IR). The short analysis time (less than 3 min) together with high reproducibility makes the newly developed method applicable to process control and screening purposes (average of the standard deviations calculated from four repetitive measurements of six different real samples: ethanol: 0.55 g L–1, glycerol: 0.037 g L–1, glucose: 0.056 g L–1, fructose: 0.036 g L–1, citric acid: 0.020 g L–1, tartaric acid: 0.010 g L–1, malic acid: 0.052 g L–1, lactic acid: 0.012 g L–1 and acetic acid: 0.026 g L–1).