Comparison of Headspace Sampling and Stir Bar Sorptive Extraction in the Detection of Whiskey Adulteration with a Mass-Spectrometry Based Chemical Sensor

Significance of the topic

Verification of authenticity is a critical aspect of food and beverage quality control. Adulteration of high-value spirits such as aged bourbons undermines consumer trust and poses economic and regulatory risks. Rapid, reliable screening methods are therefore essential to detect additions of coloring agents, diluents or lower-value products.

Objectives and Study Overview

This work evaluates the performance of two sample introduction techniques—static headspace (HS) sampling and stir bar sorptive extraction (SBSE or Twister)—in conjunction with a mass spectrometry‐based chemical sensor. The target is to distinguish unadulterated bourbons from samples spiked with caramel coloring, water or cheaper bourbon. Multivariate analysis methods (PCA, SIMCA, PCR) are used to build classification and calibration models.

Methodology and Instrumentation

• Sample preparation for HS: 5 mL bourbon equilibrated at 75 °C for 20 min in 10 mL vials, 30:1 split injection into MSD without chromatographic separation.
• Sample preparation for SBSE: 10× diluted bourbon extracted by stir bar for 1 h, rinsed and desorbed in a thermal desorption unit coupled to GC–MS.
• Instrumentation: Gerstel ChemSensor 4440 with Agilent 7694 headspace unit and 5973 MSD; Gerstel Twister with Gerstel TDS2/TDSA thermal desorber and Agilent 6890 GC–5973 MSD.
• Chemometric software: Pirouette 3.02 and Instep 2.0 for PCA, SIMCA, PCR modeling.

Main Results and Discussion

SBSE introduction provided signal intensities up to four orders of magnitude higher than static HS and revealed higher-mass ions (m/z 100–170) not seen in HS spectra. PCA models built on SBSE data captured 87 % of variance in three principal components versus 73 % for HS. SIMCA interclass distances among 1, 3 and 6 year bourbons exceeded 3 for both techniques but were consistently larger with SBSE, indicating stronger class separation. Adulteration scenarios were successfully identified: blending with cheaper bourbon was detected down to 2 % v/v, water dilution to 10 % v/v, and caramel coloring alteration was flagged by chemometric classification despite unaltered headspace composition.

Benefits and Practical Applications

Combining SBSE or HS sampling with a MS‐based chemical sensor enables rapid, non‐targeted screening of spirit authenticity. The approach discriminates closely related samples and detects low-level adulteration without lengthy chromatographic methods. It is suitable for quality control in production, regulatory surveillance and brand protection.

Future Trends and Applications

Integration of SBSE-MS sensors into automated on-line monitoring systems could further accelerate authenticity testing. Expansion to other beverages, such as wines and spirits, and development of miniaturized or portable MS instruments will broaden field deployment. Advanced chemometric algorithms and machine learning may enhance detection of novel adulterants and complex mixtures.

Conclusion

This study demonstrates that SBSE combined with mass spectrometry and multivariate analysis outperforms static headspace sampling in sensitivity and class separation for bourbon adulteration screening. Both techniques reliably detect common adulterants, supporting their use in rapid quality control workflows.

References

• Straight Bourbon, What is bourbon? (accessed March 2002).
• Bronze M.S., Vilas Boas L.F., Belchior A.P. J. Chromatogr. A 1997, 768, 143–152.
• Wiskur S.L., Anslyn E.V. J. Am. Chem. Soc. 2001, 123, 10109–10110.
• Headley L.M., Hardy J.K. J. Food Sci. 1989, 54, 1351–1354.
• Saxberg E.H., Duewer D.L., Booker J.L., Kowalski B.R. Anal. Chim. Acta 1978, 103, 201–212.
• Pfannkoch E., Whitecavage J. Pittcon 2000, Poster 2235.
• Tienpont B., David F., Bicchi C., Sandra P. J. Microcolumn Separations 2000, 12, 577–584.