Abstract
To enhance transparency in the soybean supply chain and help prevent misrepresentation of geographic origin, an analytical method combining ICP-MS with chemometrics was developed. A total of 422 soybean samples were collected from Brazil, the United States, Argentina, China, India, Paraguay and Canada, representing over 95% of global production. The OPLS-DA multivariate analysis model used for classification achieved 98.5% accuracy, with Ni, Na, Mo, Ba, Co, Cr, Cd, Sr, Se, K and Ca identified as key elements for origin differentiation. This approach provides a practical tool for companies and regulators to verify geographic origin, supporting compliance with trade and sustainability requirements and tariff-related controls. Additionally, the ability to differentiate soybean samples from various regions within Brazil and the United States was investigated and preliminary comparisons of meal samples from deforested and non-deforested areas in Brazil revealed elemental differences, suggesting potential environmental influences and highlighting the need for further investigation.