Harrison Fuller, James Harbertson,* and Christopher Beaver
*Washington State University, 359 University Drive, Richland, WA 99354 (jfharbertson@wsu.edu)

Raman spectroscopy is presented as a tool to monitor enological fermentations. Unlike the spontaneous Raman effect, fluorescence is the result of light-induced change in molecular electronic state. Fluorescence is a highly efficient process and consequently contributes much more signal than traditional Raman scattering. For this reason, the presence of fluorescence has made many instances of Raman spectroscopy applications impractical, as any spectrum generated from fluorescing compounds in a sample masks the more subtle Raman spectrum. This work employs adsorption methods to mitigate fluorescence while monitoring alcoholic fermentation. Ethanol and total sugars (fructose plus glucose) of wines made from red (Cabernet Sauvignon, Merlot, Syrah, Petit Verdot, and blends) and white (Chardonnay) cultivars were modeled using partial least squares regression (PLSR) and Ridge regression. The results comparing predicted to measured ethanol concentrations were respectable for fermenting and finished wines (R² _PLS= 0.93, R² _Ridge= 0.82) and excellent for samples stripped of phenolic contents (R² _PLS= 0.99, R² _Ridge= 0.99). Total sugar models also performed well when comparing predicted to measured values (R² _PLS= 0.87, R² _Ridge= 0.92).

Funding Support: The Wine Research Advisory Committee, the Washington Wine Commission, and the Washington Grape and Wine Research Program.