The Analysis of Beer Components Using FT-NIR Spectroscopy

Significance of the Topic

The composition and physical properties of beer determine its sensory profile, stability and regulatory compliance. Rapid, multi-parameter analytical methods that minimize sample preparation are valuable for brewery QA/QC and process control because they shorten decision cycles, reduce cost and enable inline or at-line monitoring. FT-NIR spectroscopy offers a single, nondestructive measurement that can quantify several beer attributes simultaneously, making it attractive for modern production environments.

Objectives and Study Overview

This application study evaluated the feasibility of using Fourier-transform near-infrared (FT-NIR) spectroscopy to quantify key beer parameters—alcohol content, color (EBC scale), refractive index and specific density—using a transflectance probe accessory. The goals were to demonstrate fast, preparation-free analysis and to develop robust chemometric models that correlate FT-NIR spectra with reference measurements.

Methodology

FT-NIR spectra were collected for 27 beer standard samples without sample preparation using a fiber-optic transflectance probe. Measurement parameters were: spectral range 10,000–4,000 cm^-1, spectral resolution 8 cm^-1, 32 co-averaged scans per acquisition, and Spectralon® reference background. Typical acquisition and prediction time per sample was ~25 seconds.

Spectral preprocessing included mean-centering and multiplicative signal correction (MSC) to account for pathlength and scatter effects. Specific spectral regions and treatments were selected per analyte and used to build Partial Least Squares (PLS) calibration models. Derivative and smoothing choices included first and second derivatives and Norris derivative smoothing with different segment/gap settings tailored to each parameter. Models were developed and validated using cross-validation diagnostics and Predicted Residual Error Sum of Squares (PRESS) analysis.

Used Instrumentation

• Thermo Scientific Antaris II FT-NIR Method Development Sampling (MDS) System
• Thermo Scientific SabIR fiber-optic probe with transflectance accessory
• Thermo Scientific RESULT acquisition software
• Thermo Scientific TQ Analyst chemometrics software (PLS calibration)
• Spectralon reference standard for background

Main Results and Discussion

PLS models showed excellent agreement with conventional reference techniques across all four target parameters. Reported correlation coefficients were very high (≈0.996–0.999+), indicating near-linear correspondence between FT-NIR predictions and reference values. Error metrics reported for the calibrations were low, with Root Mean Square Error of Calibration (RMSEC), Root Mean Square Error of Cross-Validation (RMSECV) and Root Mean Square Error of Prediction (RMSEP) demonstrating good predictive performance. PRESS plots exhibited the expected decline to a stable minimum, supporting model robustness and appropriate component selection.

The key practical outcomes were: rapid analysis (~25 s/sample), no chemical reagents or sample prep required, and the ability to predict multiple quality attributes from a single spectrum. The transflectance probe configuration enabled direct liquid measurements suitable for routine QA/QC and potential at-line implementation.

Benefits and Practical Applications

• Time and cost reduction versus separate reference methods (e.g., distillation, pycnometry, chromatography).
• Simultaneous multi-component quantification from one measurement streamlines QA workflows.
• Minimal operator effort and no consumable reagents reduce running costs and variability.
• Short measurement times enable higher throughput and support near-real-time process control.

Future Trends and Applications

FT-NIR with fiber-optic probes can be extended beyond the analyzed parameters to additional compositional and process indicators (e.g., residual sugars, bitterness-related compounds, higher alcohols) as robust calibration data become available. Integration with process control systems and inline probe deployments will promote closed-loop control of fermentation and blending steps. Advances in chemometric approaches, transfer learning, and standardized calibration transfer methods will ease deployment across sites and instruments. Portable and miniaturized NIR platforms may broaden adoption for field and small-batch breweries.

Conclusion

This application demonstrates that FT-NIR spectroscopy, applied with a transflectance fiber-optic probe and PLS chemometrics, is a fast, accurate and robust method for routine analysis of beer properties such as alcohol content, color, refractive index and specific density. The approach reduces analysis time and sample handling compared with conventional techniques and supports improved QA/QC and process efficiency in brewing operations.

References

• Thermo Fisher Scientific Application Note AN51892 (2021): The Analysis of Beer Components Using FT-NIR Spectroscopy, Antaris II MDS System.