Fast arsenic speciation analysis of wine and rice with LC-ICP-QQQ.

Metals and Metals Speciation Analysis in Environmental Samples
Oral Presentation

Prepared by C. Tanabe1, P. Gray2, J. Nelson3, S. Ebeler1
1 - University of California, Davis; Department of Viticulture and Enology, University of California, Davis RMI North, 595 Hilgard Lane, Davis, CA, 95616, United States
2 - Center for Food Safety and Applied Nutrition, US Food and Drug Administration, 5001 Campus Drive, College Park, MD, 20740, United States
3 - Agilent Technologies, Inc., Agilent Technologies, Inc., 5301 Stevens Creek Blvd., Santa Clara, CA, 95051, United States


Contact Information: cktanabe@ucdavis.edu; 253-632-7759


ABSTRACT

Arsenic (As) naturally occurs in the environment and is consequently found in food and beverages such as rice based products and wine. It exists in multiple forms, however, different species range in toxicity. Due to the potential health threat, some regulations have been proposed for the more toxic inorganic arsenic species (AsV and AsIII). Traditionally, these values were achieved by measuring individual species using ion exchange high pressure liquid chromatography coupled to a triple quadrupole inductively coupled plasma - mass spectrometer (HPLC-ICP-QQQ) and adding the two inorganic forms together. Instead, for this fit-for-purpose method, AsIII was intentionally oxidized to AsV with hydrogen peroxide prior to analysis, allowing all inorganic arsenic to be expressed as AsV. This allowed the inorganic As to be separated from monomethylarsonic acid and dimethylarsinic acid in 2 minutes using a narrow bore, small particle HPLC column. This analysis time is 10 times faster than the current Food and Drug Administration methods for the speciation of As. Furthermore, the use of O2 as a reaction gas in the ICP-QQQ allowed for a decrease in spectral interferences while increasing sensitivity. A small injection volume also helped to mitigate non-spectral interferences such as carbon enhanced ionization. The validation results from two participating laboratories is presented to demonstrate the new methodís accuracy and reproducibility in wine and rice matrices.