In this paper, Cr (Ⅲ) and Cr (Ⅵ) in milk were determined by high performance liquid chromatography coupled with inductively coupled plasma mass spectrometry (HPLC-ICP-MS). A PRP-X100 cation exchange column was used. The influence of mobile phase concentration, pH value and flow rate on the degree of separation and retention time was investigated. The mobile phase selected was NH4NO3solution (0.10 mol/L, pH 7) at a flow rate of 1.0 mL/min. Under the optimized conditions, the mass spectrometric and ion interferences were eliminated by using the natural abundance ratio of 52Cr to 53Cr as reference value, and the retention time of Cr (Ⅲ) and Cr (Ⅵ) were determined to be 2.227 and 1.856 min, respectively. Good linearity was observed for Cr (Ⅲ) and Cr (Ⅵ) with correlation coefficients of 0.999 5 and 0.999 7, respectively. The detection limits were 0.015 and 0.016 μg/L, respectively. The overall recoveries for the two valence states were in range of 71.52% to 90.27%. This method proved to be useful for simultaneous analysis of different Cr valence states in milk and dairy products.

Inductively coupled plasma mass spectrometry (ICP-MS) and inductively coupled plasma atomic emission spectrometry (ICP-AES) were compared for determining the contents of metal elements in infant formula milk powder. The limits of detection (LOD) and quantitation (LOQ) of ICP-MS were established using Sc, Rh and Bi as internal standards. Relative standard deviations (RSDs) for 26 analytes were less than 5.0%, and recoveries from spiked samples were 82.80%-110.80%. Eight metal elements could be simultaneously determined by ICP-AES, whose LOD and LOQ were slightly higher than those of ICP-MS. Furthermore, ICP-AES exhibited a good linearity and spiked recoveries in the range of 91.64%-112.73%. In the end, we concluded that both ICP-MS and ICP-AES had a high precision and a good linearity, but ICP-MS allowed the determination of a wider range of metal elements with lower LOD and LOQ.