Determining the geographical and botanical origins of honey by harnessing the power of triple quadrupole ICP-MS technology

Importance of the topic

Honey is a high-value natural product prone to adulteration and mislabeling of botanical and geographical origin. Reliable origin assignment supports consumer confidence, food safety, regulatory compliance, allergy management, biodiversity preservation and sustainable beekeeping.

Objectives and overview

This study presents a complete analytical workflow combining automated sample preparation, triple quadrupole ICP-MS analysis and multivariate statistics to predict the botanical and geographical origin of over 120 honey samples.

Methodology and instrument

• Sample selection: 122 honeys with certified monofloral and regional origins (Europe, Asia, South America).
• Sample preparation: duplicate microwave-assisted acid digestion (ultraWAVE system with EasyFILL dispenser) of 1.0 g honey; HNO₃, HCl and H₂O₂ reagents; fully automated workflow.
• Instrumentation: Thermo Scientific iCAP TQ ICP-MS with SC-4DX autosampler and FAST valve; MicroMist nebulizer; quartz cyclonic spray chamber; Ceramic Torch Plus; modes: SQ-NoGas, SQ-KED (He), TQ-O₂.
• Software: Qtegra ISDS for data acquisition; Minitab Statistical Software for model building and prediction.
• Calibration: multi-element standards covering major (mg/L) and trace (µg/L) elements; internal standards Ge, In, Rh, Ir for drift correction; robust linearity and low detection limits.

Main results and discussion

• EU vs Non-EU model: binary tree using Na, P, Cr, K; 2 % classification error; predictive accuracy on 19 unknown samples ≥ 90 %.
• France vs China model: two-element tree based on P (as P-O₁₆) and Na; distinct clusters in score plots; 100 % correct origin assignment for 20 test samples.
• Botanical origin model: multinomial tree with seven elements (B, K, Mg, Ca, Ba, Co, Al); overall misclassification 7.3 %; clear clusters for acacia, lavender, linden, sunflower, chestnut, rapeseed and mixed honeys.

Benefits and practical applications

• High throughput and reproducibility through automation minimizes contamination and operator bias.
• Simultaneous quantification of major and trace elements in a single run with effective interference removal.
• Scalable database supports routine authenticity testing in QC laboratories and enforcement agencies.
• Statistical models provide probabilistic origin assignments to detect fraud and mislabeling.

Future trends and possibilities

• Integration of isotopic ratio analysis on the same ICP-MS platform for complementary authenticity markers.
• Application of advanced machine learning algorithms for more complex origin and adulteration detection.
• Extension of the workflow to other food matrices and environmental samples.
• Development of portable or benchtop triple quadrupole ICP-MS systems for on-site authenticity screening.
• Establishment of shared international databases to harmonize origin certification standards.

Conclusion

This work demonstrates a robust, automated triple quadrupole ICP-MS strategy for reliable prediction of honey origin. The combination of precise elemental fingerprinting, interference-free analysis and statistical modelling yields high accuracy in both botanical and geographical classification. The workflow is readily transferable to routine laboratories to enhance food authenticity control.

References

1. EU Coordinated Action on honey fraud; Analytical testing results of imported honey; Official Journal of the European Communities L10 (2002).
2. European Commission 2001/110/EC relating to honey, Official Journal of the European Communities L10, 47–52 (2002).
3. Kracht O., Hilkert A. EA-IRMS Detection of honey adulteration, Thermo Fisher Scientific Application Note 30177 (2016).