Arsenic speciation in commercially available peanut butter spread by IC-ICP-MS

Abstract

Six commercially available peanut butter spread samples were evaluated for total arsenic content, inorganic and organic species. Total arsenic values (ranging from 6.5 ng g−1 to 21.4 ng g−1) were determined after heated digestion with nitric acid–hydrogen peroxide using methanol-enhanced standard nebulization ICP-MS. Three percent methanol was added to all standards, blanks and sample digestates for enhancement of the arsenic signal. For arsenic speciation, three of the products (JCPB1, TJPB1 and SCPB1) were chosen based on their total arsenic values and because each one represented a different formulation (smooth, crunchy and organic). Various organic solvent–water mixtures were evaluated as the extraction solution. After several failures, based on <10% spike recoveries, an alternative solvent (2-butoxyethanol) was tried. This solvent had been successful in solvating various drug products for total metals determination by ICP-MS. By adjusting the time in the sonicator and optimizing the time/temperature in the reciprocating water bath, spike recoveries (40–122%) could be achieved. Four arsenic species were determined in one of the products (JCPB1) studied, the values were: As+3 = 1.6 ng g−1; DMA = 6.1 ng g−1; MMA = 4.9 ng g−1; As+5 = 6.3 ng g−1. The sum total of the four species determined equaled 19 ng g−1, compared to the total arsenic at 19.0 ng g−1. A duplicate sample was carried through at the same time and the total speciated arsenic = 18.2 ng g−1. Total arsenic values (non-certified) for a new SRM 2387, reported for the first time, were determined to be 12 ng g−1, while two organoarsenic species (DMA and MMA) were determined to be on average 8.3 and 4.7 ng g−1, respectively. Further, the IC-HG-ICP-MS mode was used to verify that the arsenic species determined in the matrix were essentially free from interferring analytes (no overlapping analytes detected).