A NON-TARGETED APPROACH TO THE DEVELOPMENT OF A FOOD ADDITIVE CCS SCREENING LIBRARY AND ITS APPLICATION

Importance of the topic

Food additives play a crucial role in modern food processing and safety, but comprehensive screening remains challenging due to the diversity of compounds and food matrices. Incorporating collision cross section data from ion mobility spectrometry into high throughput liquid chromatography mass spectrometry workflows offers enhanced specificity and reduced false detections. Developing a broad, non-targeted library of additives supports regulatory compliance and protects consumer health.

Objectives and overview of the study

This work aimed to establish a universal, non-targeted UPLC-MS ion mobility method to build a screening library of food additives. The library includes sweeteners, preservatives, antioxidants and colorants. Seven commercial food and drink products labelled as containing additives were analyzed to assess the robustness of the collision cross section enhanced library in both positive and negative ion modes.

Methodology and sample preparation

Samples including yogurts, soft drinks, energy and tonic beverages were extracted with acidified acetonitrile and magnesium sulfate based cleanup. Dispersive solid phase extraction was performed with PSA and C18 sorbents. Extracts were diluted for liquid chromatography mass spectrometry analysis. Data acquisition used HDMSE mode at 10 spectra per second, applying a collision energy ramp for fragmentation and travelling wave ion mobility separation in nitrogen buffer gas.

Used Instrumentation

• UPLC system: Waters ACQUITY UPLC I-Class
• Columns: ACQUITY UPLC BEH C18 and HSS C18
• IMS-TOF MS: Waters Synapt G2-Si with travelling wave ion mobility
• Software: UNIFI 1.9.2 for data processing and CCS calibration

Main results and discussion

The screening library comprises precursor ions, fragment ions and collision cross section values for over forty food additives. Routine CCS delta values were below two percent in both ionization modes. In lemon and zero sugar drinks, sweeteners such as neohesperidin, acesulfame and sucralose and preservatives like citric acid were correctly identified, with no false positives. Illegal additives were spiked into a tonic drink and successfully distinguished within CCS tolerance. Ion mobility resolved conformers and protomers for compounds like tartrazine, revealing two CCS values and corresponding fragmentation spectra.

Benefits and practical applications

• Broad coverage of additive classes in a single non-targeted assay
• Enhanced specificity through CCS criteria to reduce false detections in complex matrices
• Rapid screening suitable for regulatory and quality control environments
• Capability to detect charged isomers and conformers not resolved by mass alone

Future trends and opportunities

Expanding collision cross section libraries to include emerging additives and metabolites will further improve screening power. Combining ion mobility data with machine learning could automate identification and distinguish closely related isomers. Cross platform CCS standardization and long-term robustness studies will support wider adoption in food safety laboratories.

Conclusion

A non-targeted UPLC-IMS-MS approach successfully generated a comprehensive food additive library incorporating collision cross section metrics. The method demonstrated high specificity and robustness across diverse food matrices, with potential to streamline regulatory screening and improve consumer protection.

Reference

1. Regulation EC No 882 2004 on official controls in food chain
2. Directives 94 35 CE, 94 36 CE and 95 2 CE on food additives
3. Goscinny S, McCullagh M Applications of CCS in pesticide screening ASMS 2013
4. McCullagh M, Goscinny S Waters Appl Note 2014 720005028E
5. Goscinny S et al Rapid Commun Mass Spectrom 2019 1 15
6. McCullagh M et al ASMS 2019 Atlanta
7. McCullagh M et al Rapid Commun Mass Spectrom 2019 1 11
8. Pereira CAM et al Phytochemical Analysis 2019
9. McCullagh M et al Anal Chem 2018 90 4585 4595
10. Goshawk J, Barknowitz G, McCullagh M ASMS 2019