Packaging Contaminants in the Headspace of Commercially Available Cola Beverages

Significance of the topic

The migration of chemical constituents from packaging into food and beverages can alter product quality, change sensory profiles, and pose potential health risks. Non-targeted volatile and semi-volatile screening of packaged foods provides early detection of unexpected contaminants (e.g., plasticizers, monomers, oligomers, and processing residues) and supports regulatory compliance with frameworks such as EU 1935/2004 and EU 2020/1245. Rapid, reliable analytical workflows that combine comprehensive separation with advanced data alignment accelerate identification and trend analysis across multiple commercial products.

Objectives and study overview

This application study used GC-TOFMS combined with ChromaTOF Sync software to perform a non-target screening for packaging-derived contaminants in the headspace of thirteen commercially available cola beverages from eight brands. The original aim was to survey aroma compounds; during data review, unexpected packaging-related analytes were discovered and characterized. The study demonstrates both the analytical power of GC-TOFMS for non-target work and the practical advantages of data alignment and visualization for multi-sample comparisons.

Methodology

Sample set: thirteen commercial cola beverages (including multiple unsweetened formulations) and a method blank were analyzed. Sample preparation and detailed instrumental parameters are reported in the instrument table of the original note; headspace extraction (HS-SPME) and pooled evaporative approaches were used as indicated in the keywords.

Data acquisition and processing: GC coupled to a time-of-flight mass spectrometer (Pegasus BT) acquired comprehensive full-scan MS data. Initial per-sample peak finding in ChromaTOF (signal-to-noise threshold 10) returned ~400 peaks per sample. To enable efficient multi-sample comparison, ChromaTOF Sync was used for automated peak finding, retention-time alignment, and compilation into a single combined peak table (~600 unique features across all samples). The software provided heat maps, per-analyte trend plots, spectrum/library matching, and multivariate analysis (PCA). Retention index (RI) confirmation and spectral similarity scores were used for tentative identifications.

Instrumentation used

• GC-TOFMS: Pegasus BT (LECO Corporation).
• Data processing: ChromaTOF peak finding and ChromaTOF Sync 1D for data alignment, heat maps, and PCA visualization.
• Sample introduction: Headspace solid-phase microextraction (HS-SPME) with pooled evaporative headspace approaches referenced in keywords and methods table.

Results and discussion

• Complex chromatograms were obtained for all cola samples, with hundreds of volatile/semi-volatile features per sample. Using ChromaTOF Sync reduced the analyst workload by consolidating features across samples into a single aligned peak table (~600 features total instead of ~5200 separate per-sample peaks).
• Packaging-related analytes tentatively identified at elevated levels (relative to the method blank) included carboxylic acids (e.g., 2-ethylhexanoic acid, 3,5,5-trimethylhexanoic acid), phthalates (diethyl phthalate, dibutyl phthalate), and styrene.
• Specific observations:
  • 2-Ethylhexanoic acid and 3,5,5-trimethylhexanoic acid were notably elevated in Cola 5 and Cola 6 (both unsweetened products from the same manufacturer), indicating a likely common packaging or processing source.
  • Diethyl phthalate and dibutyl phthalate were tentatively identified in all cola beverages; spectral similarity scores (~805 and 881) and RI confirmation (observed RI 1593 vs library 1594 for diethyl phthalate; observed RI 1960 vs library 1965 for dibutyl phthalate) supported the assignments. Higher relative abundances were observed in Cola 2, Cola 6, and Cola 11.
  • Styrene was tentatively elevated in Cola 7 (spectral similarity ~821; observed RI 897 vs library 893). Styrene is a known packaging contaminant with established exposure guidance; the detection here flags the need for targeted quantification to compare against health-based limits.
• ChromaTOF Sync effectively handled small retention-time shifts across the dataset (example: 2-ethylhexanoic acid shifted by ~1.2 seconds between samples) and automated the matching process, reducing manual review time.
• Multivariate analysis (PCA) of the aligned peak table separated the thirteen beverages along principal components, revealing clustering patterns: Cola 3 and Cola 4 (both “original” formulations from different companies) showed high chemical similarity, and unsweetened variants exhibited greater profile divergence compared with sweetened counterparts.

Benefits and practical applications

• The GC-TOFMS + ChromaTOF Sync workflow enables efficient non-target surveillance of packaging migrants across multiple consumer products, combining sensitivity, deconvolution, and automated alignment to streamline data interpretation.
• Heat maps and per-analyte trend plots expedite identification of sample-specific elevations and possible sources (manufacturer, packaging material, or formulation differences).
• Early detection of potential contaminants (phthalates, styrene, plasticizer-related acids) informs risk assessment and prioritizes analytes for subsequent targeted quantification and regulatory evaluation.
• The approach supports quality control, forensic investigations of off-flavors, supplier audits, and compliance with food-contact regulations when followed by validated quantitative methods.

Future trends and potential applications

• Follow-up work should focus on validated, targeted quantitative methods for prioritized migrants (calibration, limits of detection/quantification, matrix-matched standards) to assess compliance with regulatory thresholds and exposure limits.
• Integration of high-resolution MS/MS libraries and orthogonal techniques (e.g., LC-HRMS for non-volatile migrants) will broaden the scope of non-target screening and improve confident identification.
• Automated pipelines combining alignment, suspect screening libraries, and semi-quantitative reporting will accelerate routine monitoring across larger product sets and supply chains.
• Development of standardized workflows and inter-laboratory studies would help translate non-target findings into enforceable limits and guidance documents for food-contact materials.

Conclusion

This study demonstrates that GC-TOFMS combined with ChromaTOF Sync is an effective platform for non-target screening of packaging-derived contaminants in beverage headspace. The workflow reduced manual review burden through automated peak alignment and visualization, uncovered recurrent packaging migrants (carboxylic acids, phthalates, styrene), and revealed product-specific patterns influenced by manufacturer and formulation. These findings warrant targeted quantitative follow-up to determine absolute concentrations and assess compliance with regulatory and health-based guidance.

References

1. Graíño SG, Sendón R, Hernández JL, Rodríguez-Bernaldo de Quirós A. Polymers. 2018;10:802. DOI: 10.3390/polym10070802.
2. Sadighara P, Akbari N, Mostashari P, Yazdanfar N, Shokri S. Food Chemistry. 2022;100238:1-4. DOI: 10.1016/j.fochx.2022.100238.
3. Chiesa LM, Soncin S, Panseri S, Cantoni C. Veterinary Research Communication. 2008;32:S319–S321. DOI: 10.1007/s11259-008-9138-7.
4. Holmes MJ, Hart A, Northing P, Oldring PKT, Castle L, Stott D, Smith G, Wardman O. Food Additives & Contaminants. 2022;22(10):907-919. DOI: 10.1080/02652030500307172.