Optimization of sample preparation of Brazilian honeys for TQ-ICP-MS analysis

• Published in: Talanta open 2022-08, Vol.5, p.100117, Article 100117
• Main Authors: Luccas, Fernanda S., Fernandes, Elisabete A. De Nadai, Mazola, Yuniel T., Bacchi, Márcio A., Sarriés, Gabriel A.
• Format: Article
• Language: English
• Subjects: Apis mellifera sp; Chemometrics; Honey; Stingless bee; Tetragonisca angustula (Jataí)
• ISSN: 2666-8319; 2666-8319
• DOI: 10.1016/j.talo.2022.100117

•Optimization of honey microwave-assisted digestion.•TQ-ICP-MS and NAA assessed chemical profile of Brazilian honeys.•ML algorithms discriminated honeys by entomological origin with 99% accuracy. Mass spectrometry-based techniques have been used to study the chemical profile of honeys to authenticate entomological, botanical and geographical origins. Sample preparation is a crucial step of the analysis to obtaining reliable data and minimizing interference owing to matrix effects. The present work studied the best sample digestion procedure for elemental analysis of Brazilian honeys from Tetragonisca angustula (Jataí) and Apis mellifera sp (Apis) by triple quadrupole inductively coupled plasma mass spectrometry (TQ-ICP-MS). A central composite design with 2² factorial and 3 center points considering different volumes of HNO3 and H2O2 was investigated. There was no statistically significant influence of the amounts of HNO3 and H2O2 on the recoveries of Ag, Al, As, Ba, Be, Ca, Cd, Ce, Co, Cr, Cs, Cu, K, La, Mg, Mn, Na, Ni, Pb, Rb, Se, Sr, Th, U, V and Zn mass fractions. Machine learning algorithms (Multilayer Perceptron, Random Forest and Support Vector Machine) allowed discriminating entomological origin of honeys based on chemical profile with a classification accuracy of 99%. [Display omitted]