Advanced profifiling of food volatiles by GC×GC-qMS: a comparative analysis of roasted hazelnuts from different origins

Significance of the Topic

Comprehensive volatile profiling is essential in the food industry for authenticating origin, monitoring quality and optimizing flavor. Complex matrices such as roasted nuts contain hundreds of volatile compounds that require advanced separation and sensitive detection to be fully characterized. Two-dimensional gas chromatography coupled with quadrupole mass spectrometry (GC×GC-qMS) addresses this need by delivering unmatched resolution and detailed chemical fingerprints, enabling reliable classification and fingerprinting of food products.

Objectives and Overview

This study aimed to compare and discriminate the volatile profiles of roasted hazelnuts from different geographical sources. Two complementary data-processing strategies were evaluated: chromatographic fingerprinting based on region features in the 2D space, and comprehensive template matching using individual peaks and MS spectral match factors. The goal was to assess their ability to reveal origin-related differences and support authenticity assessments.

Methodology and Instrumentation

• Sample preparation by headspace solid-phase microextraction (HS-SPME) of roasted hazelnut volatiles.
• GC×GC-qMS system: Agilent 6890 gas chromatograph, Zoex KT 2004 loop thermal modulator, Agilent 5975 MSD detector.
• First- and second-dimension columns coupled in sequence to boost separation capacity.
• Data visualization and processing with Zoex GC Image software, including Image Investigator for advanced fingerprinting.

Main Results and Discussion

• Chromatographic fingerprinting generated a cumulative 2D chromatogram of nine hazelnut samples to locate distinctive regions; polygonal windows captured region-based features for classification.
• Comprehensive template matching extracted approximately 450 untargeted 2D peaks into a master template, using retention coordinates and MS match factors to align peaks across samples.
• Fingerprinting highlighted global pattern differences and offered rapid screening of significant 2D regions, showing clear grouping by origin.
• Template matching provided enhanced specificity, assigning match percentages to sample pairs and enabling quantitative comparison of individual aroma compounds.
• Both approaches yielded consistent discrimination of hazelnut varieties from Italy, Germany and the USA, supporting the robustness of GC×GC-qMS profiling.

Benefits and Practical Applications

• Authentication of geographical origin and detection of adulteration in high-value nuts.
• Quality control during roasting processes by monitoring key volatile markers.
• Formulation support for flavor development in nut-based products.
• Data-driven optimization of supply chains through chemometric grouping.

Future Trends and Possibilities

• Integration with machine-learning algorithms for automated pattern recognition and prediction.
• Expansion of comprehensive 2D libraries to cover diverse food matrices.
• Coupling GC×GC with complementary detectors such as GC-infrared spectroscopy for structural confirmation.
• Development of portable or miniaturized GC×GC platforms for in-line quality monitoring.
• Real-time online analysis in production environments for rapid decision-making.

Conclusion

GC×GC-qMS combined with advanced data-processing techniques offers a powerful toolset for in-depth profiling of complex food volatiles. Chromatographic fingerprinting and template matching each contribute unique strengths, together ensuring sensitive, specific and reproducible discrimination of roasted hazelnut origins. This approach enhances authenticity testing and flavor optimization in the food sector.

Reference

• JSB Author Application Note: Daniela Peroni, JSB © 2015