Semi-quantitative analysis of nutrients in milk-based powders by EDXRF and UniQuant software

Importance of the Topic

Milk-based powders, such as infant formula, require reliable nutrient profiling to ensure product quality and compliance with nutritional standards.
Energy-dispersive X-ray fluorescence (EDXRF) offers rapid, near-line elemental analysis, reducing turnaround time and resource requirements in food production.

Objectives and Overview

This application note demonstrates the use of Thermo Scientific UniQuant standardless fundamental parameters software with the ARL QUANT'X EDXRF spectrometer to perform semi-quantitative analysis of major and trace nutrients in fortified milk powders.
Comparison against reference methods (ICP-OES and potentiometry) evaluates the accuracy and precision of the standardless approach.

Methodology and Used Instrumentation

Instrument configuration:

• Thermo Scientific ARL QUANT'X EDXRF spectrometer with 50 W Ag-target X-ray tube and graphene-window SDD detector
• Vacuum measurement geometry with nine selectable excitation conditions and optimized filters to cover elements from carbon (Z = 6) to americium (Z = 95).

Sample preparation and measurement:

• Approximately 6.0 ± 0.1 g of powder pressed into 32 mm pellets using a manual hydraulic press at 4 tons, without grinding or drying.
• Measured on both sides to ensure sample homogeneity, with a total live time of 720 s across six key excitation conditions targeting Na through Zn.

Calibration and software approach:

• UniQuant fundamental parameters model calibrated with pure elements, compounds, and reference materials for universal semi-quantitative analysis (relative error ~15–30 %).
• Application-specific optimization of sensitivity factors using 13 in-house milk powder standards characterized by ICP-OES and potentiometry.

Main Results and Discussion

Validation against 23 in-house standards measured in duplicate yielded correlation coefficients (R²) between 0.87 and 0.98, with slopes close to unity for most elements.
Standard error of prediction (SEP) values indicated acceptable precision, though confidence intervals for intercepts revealed minor biases.
Overall scatter was higher than that observed with empirical calibrations, but the standardless method reliably estimates concentrations without requiring extensive reference standards.

Benefits and Practical Applications

• Eliminates the need for extensive calibration standards, simplifying workflow and reducing costs.
• Enables rapid, near-line monitoring of key nutrients, supporting real-time quality control in production environments.
• Detects a broad range of essential macro- and trace elements, including light elements down to magnesium.
• Applicable for routine QA/QC, regulatory compliance, and forensic investigations of milk-based products.

Future Trends and Opportunities

Advancements in detector materials and window technologies may enhance sensitivity for light elements.
Refinement of fundamental parameters models, incorporation of machine learning algorithms, and automated calibration routines are expected to reduce bias and improve accuracy.
Integration with process analytical technology (PAT) frameworks and digital twins could further streamline production monitoring and control.

Conclusion

The combination of ARL QUANT'X EDXRF spectrometry and UniQuant standardless fundamental parameters software provides a practical semi-quantitative solution for nutrient analysis in milk-based powders.
While not matching the precision of empirical calibration, its flexibility and rapid turnaround make it an attractive tool for food industry applications.

References

No external references were provided in the original text.