Dual parallel detection raw data fusion: Challenges and opportunities for accurate fingerprinting over large time frames

Importance of the topic

Comprehensive GC×GC with dual detection (MS and FID) is critical for the analysis of complex mixtures in food, fragrance, and cosmetic industries. Combining molecular fingerprinting with robust quantification enhances sample discrimination and ensures reliable quality control over extended time frames.

Objectives and Study Overview

This work introduces a workflow to fuse raw parallel signals from MS and FID channels for accurate chromatographic fingerprinting across large datasets and long-term studies. Case applications include quantitative profiling of hazelnut volatiles and regulated allergens in raw fragrances.

Methodology and Instrumentation

Data processing begins with independent preprocessing of MS and FID chromatograms, including baseline correction, 2D peak detection, and alignment. An automated script generates a fused image by encoding the FID trace as an m/z fragment, enabling joint peak recognition based on unique retention coordinates and spectral metadata.

Used Instrumentation

• Comprehensive GC×GC system with dual parallel detectors
• Time-of-flight mass spectrometer (MS) with 70 eV and 12 eV electron ionization
• Flame Ionization Detector (FID)
• Modulator for second-dimension separation
• Custom scripting tools for image-based data fusion and pattern recognition

Main Results and Discussion

Fusion of MS and FID data increased reliable peak matching from 20 (MS alone) and 96 (FID alone) to a consolidated dataset, reducing false positives and negatives. The fused approach enabled consistent feature templates and improved classification accuracy in 30-sample datasets. In hazelnut studies, 45 volatile markers were quantified, including lipid oxidation products, key aroma compounds, and defect indicators. Fragrance profiling achieved precise quantification of about 60 regulated allergens, meeting EU cosmetic regulations.

Benefits and Practical Applications

• Enhanced specificity through combined spectral and retention data
• Improved quantitative linearity and temporal stability
• High-throughput untargeted and targeted fingerprinting
• Compliance with regulatory standards in cosmetics
• Robust discrimination of sample origin, variety, and storage conditions

Future Trends and Opportunities

• Integration of machine learning for automated feature extraction
• Real-time data fusion workflows for on-line quality monitoring
• Expansion to additional detector combinations (e.g., olfactometry, atomic spectroscopy)
• Development of open data standards for fused chromatographic datasets
• Application to environmental analysis, biofluids, and industrial process control

Conclusion

The presented dual-channel fusion workflow successfully combines the strengths of MS and FID in GC×GC, delivering accurate fingerprinting, robust quantification, and improved sample classification. This approach addresses long-term stability challenges and opens pathways to advanced analytical applications in various sectors.

References

• Regulation (EC) No 1223/2009 on cosmetic products, European Commission, November 2009
• Commission Regulation (EU) 2023/1545, July 2023
• De Saint Laumer J.Y., Cicchetti E., Merle P., Egger J., Chaintreau A., Analytical Chemistry, 2010, 82(15), 6457–6462