A protocol based on microwave digestion with oxidizing agents (HNO
3, H
2O
2) followed by detection with sector field inductively coupled plasma mass spectrometry (SF ICP MS) was developed and in-house
validated to measure As, Cd, and Pb content in milk powder. The limits of detection (LoD) and quantification (LoQ) were 3.04
and 9.12 ng g
−1 for As, 0.49 and 1.46 ng g
−1 for Cd, 0.62 and 1.85 ng g
−1 for Pb. The signal response was linear up to two orders of magnitude of concentration. The average recovery from the certified
skim milk powder reference material was 100.6% for Cd and 97.6% for Pb, whereas spiked samples provided a mean recovery of
102% for As. Repeatability and reproducibility respectively were equal to 3.5 and 4.3% for As, 4.9 and 5.0% for Cd, and 5.7
and 5.1% for Pb. Validation data were used to calculate the method’s uncertainty, and relative expanded uncertainties were
8.6% for As, 12% for Cd and 10% for Pb. The method was robust when critical procedural parameters were changed and when applied
to cow’s milk and infant formulas. Arsenic level was <LoD in all samples, Cd ranged 2.3–4.5 ng g
−1 in cow’s milk and 2.3–7.8 ng g
−1 in infant formulas, and Pb ranged 8.9–15.9 ng g
−1 in cow’s milk and was <LoQ-12.1 ng g
−1 in infant formulas.
Keywords: Metals; milk; method validation; measurement uncertainty; SF ICP MS
Correspondence: Beatrice Bocca, Department of Environment and Primary Prevention, Istituto Superiore di Sanità, Viale Regina
Elena 299, 00161 Rome, Italy