Applying statistical data processing tools for GC×GC differentiation of alternative aviation fuels

Significance of the topic

Comprehensive two-dimensional gas chromatography coupled with time-of-flight mass spectrometry (GC×GC-TOFMS) and statistical data processing addresses the growing need to characterize and differentiate complex alternative aviation fuels. Accurate profiling of hydrocarbon classes and minor components supports regulatory compliance, quality control, and feedstock optimization in the aviation industry.

Objectives and study overview

The study aimed to apply advanced statistical tools to GC×GC data to:

• Differentiate traditional and synthetic aviation fuels based on chemical composition
• Identify clustering patterns among diverse alternative fuel production methods
• Pinpoint specific additive components in fuel blends using minimal sample sets

Methodology and Applied Instrumentation

Samples of traditional Jet A, AVTUR, Synthetic Paraffinic Kerosene (FT-SPK), HEFA-SPK and other alternative fuels were analyzed using comprehensive two-dimensional gas chromatography with a Pegasus BTX benchtop time-of-flight mass spectrometer. Key instrument parameters included a dual-stage reflectron, mass range of 10–1500 m/z, resolution >1100 at m/z 219 and acquisition rates up to 500 spectra/sec. Data processing employed LECO’s ChromaTOF software suite with modules for full acquisition control, non-target deconvolution, target analyte finding, spectral analysis and quantitation. The ChromaTOF TILE tool was used to generate tiled chromatograms and compile aligned feature tables.

Statistical comparisons included:

• Fisher Ratio (FR) for multi-class discrimination requiring replicates
• Coefficient of Variation (CoV) for evaluating within-class variability and enabling principal component analysis
• Fold Change (FC) for rapid two-sample comparisons, reported as log2 values

Main results and discussion

Fisher Ratio analysis highlighted key components distinguishing traditional and synthetic fuels, reinforcing known differences in paraffinic and aromatic content. CoV-based PCA revealed distinct clustering of fuels produced by low- vs. high-temperature Fischer-Tropsch, HEFA, ATJ, SIP and catalytic hydrothermal processes. Fold-change comparison of a pair of samples—with and without an aromatics additive—rapidly isolated peaks unique to the additive package. Overall, statistical tiling enhanced detection of low-abundance species and improved confidence in class assignments.

Benefits and practical applications of the method

The combined GC×GC-TOFMS and statistical workflow offers:

• High-resolution separation and sensitive detection of fuel components
• Objective, reproducible differentiation of fuel types for QA/QC
• Rapid identification of additive formulations with minimal sample numbers
• Data-driven support for process optimization and regulatory reporting

Future trends and potential applications

Emerging developments may include:

• Integration of machine learning algorithms for automated feature classification
• Expansion of high-throughput GC×GC platforms for real-time process monitoring
• Coupling to complementary detectors (e.g., olfactometry) for sensory profiling
• Standardization of statistical workflows to support cross-laboratory data sharing

Conclusion

The application of statistical tools to comprehensive GC×GC-TOFMS data enables robust differentiation and characterization of alternative aviation fuels. By selecting appropriate metrics—Fisher ratio, coefficient of variation or fold change—analysts can tailor workflows to specific questions, from broad class separation to targeted additive identification. This approach enhances both research insights and quality-control practices in advanced fuel analysis.

Reference

• Kelly C. Applying statistical data processing tools for GC×GC differentiation of alternative aviation fuels. Multidimensional Chromatography Workshop, University of Liège, February 2025.