On the certification of cadmium at trace and ultratrace levels in standard reference materials using ID ICP-MS

Analytical methods used for the isotope dilution inductively coupled plasma mass spectrometric (ID-ICP-MS) measurement of Cd at μg kg⁻¹ and sub-μg kg⁻¹ levels are described and applied to the certification of new dietary supplement, blood, and serum Standard Reference Materials (SRMs). The materials are: SRM 3240 Ephedra sinica Stapf Aerial Parts, SRM 3241 Ephedra sinica Stapf Native Extract, SRM 3243 Ephedra-Containing Solid Oral Dosage Form, SRM 3244 Ephedra-Containing Protein Powder, SRM 966 Toxic Metals in Bovine Blood, Level 1 (L1) and Level 2 (L2), and SRM 1598a Animal Serum. The concentration of Cd in the materials ranges from 120 μg kg⁻¹ down to 0.03 μg kg⁻¹. At these levels, the factors that most influence the accuracy of the ICP-MS data are the procedure blank and spectral and nonspectral interferences. Nonspectral interference, caused by the high concentration of dissolved solids in the matrices investigated, resulted in signal suppression. Matrix separation was used to enhance signal intensity and to reduce spectral interference for the accurate determination of Cd in SRM 1598a and SRM 3244. Chromatographic separation procedures using Chelex for SRM 1598a and anion exchange for SRM 3244 were optimized to achieve the desired separation characteristics without substantially increasing the procedure blank. Sensitivity for the determination of Cd in serum was additionally enhanced through the use of desolvation nebulization. We determined that separations were not required for the accurate ICP-MS determination of Cd in SRM 3240, SRM 3241, SRM 3243, and SRM 966 L2 under optimized analysis conditions. These samples were diluted to a minimum volume and introduced to the ICP-MS via low flow (40–100 μL/min) microconcentric nebulizers. SRM 966 L1 was also analyzed directly, but results were highly variable. The ID-ICP-MS sample preparation and ratio measurement protocols described here resulted in total expanded uncertainties of less than 1% for the determination of 90.85 μg kg⁻¹ Cd in SRM 3240, and less than 10% total expanded uncertainty for the determination of 0.0468 μg kg⁻¹ Cd in SRM 1598a.