Bright Path for Sensors - Handheld Technology for Quality Control in the Brewing Industry

Effective quality control is essential for brewers to ensure product consistency and compliance across batches, especially with the increasing diversity of beer styles in today's market. However, current beer quality evaluation methods often involve multiple equipment, time-consuming process and specialized instruments, which can pose challenges, especially for microbreweries in conducting comprehensive quality analysis. We have studied miniaturized sensors based on vibrational spectroscopy (NIR, FTIR and Raman) a rapid and high-throughput tools for the non-destructive monitoring of key quality parameters and flavor-active compounds.

We have partner with local microbreweries (Columbus, OH, USA) to collect a diverse collection of beer products, including various Lager and Ale styles. Samples were analyzed for key quality parameters—including specific gravity, real extract, alcohol content, bitterness, and color—using official methods from the American Society of Brewing Chemists. Targeted flavor-active compounds, including acetaldehyde, acetic acids, ethyl acetate, diacetyl, phenylethyl alcohol, isoamyl alcohol, furfuryl alcohol, and furfural, were quantified using gas chromatography with flame ionization detection (GC-FID). Beer spectral data were collected using handheld FT-NIR and Raman devices and a portable FT-IR unit equipped with a triple-reflection diamond ATR. Partial least squares regression (PLSR) models were developed by combining reference data with spectral data, with leave-one-out cross-validation and external validation to assess performance.

Prediction models using our handheld FT-NIR showed excellent predictive accuracy for key quality parameters, with correlation coefficients (R²) ranging from 0.90 to 0.99 and low standard errors of prediction. Volatile compound concentrations varied widely across samples: acetaldehyde (1.1–9.7 ppm), acetic acids (203–1003 ppm), ethyl acetate (0.3–29.8 ppm), phenylethyl alcohol (7–74 ppm), isoamyl alcohol (26–128 ppm), furfuryl alcohol (100–1170 ppm), and furfural (6–53 ppm). Based on these reference results, robust PLSR models were built from FT-IR spectra with R² values ranging from 0.81 to 0.96. Raman allowed to collect quality information through bottles, without opening the containers.

Our findings highlight the potential of miniaturized infrared spectroscopy, combined with multivariate data analysis, as a powerful tool for rapid and cost-effective on-site detection of quality anomalies in beer. This approach enables brewers to optimize production processes and ensure consistent product quality, offering significant benefits to both microbreweries and larger operations.