Application of hydride generation - microwave plasma - atomic emission spectrometry and partial least squares regression for the determination of antimony directly in water and in PET after alkaline methanolysis

• Published in: Chemosphere (Oxford) 2023-02, Vol.313, p.137316-137316, Article 137316
• Main Authors: Magana-Maldonado, Luis Mario, Wrobel, Katarzyna, Espinoza Cruz, Tania Lizeth, Yanez Barrientos, Eunice, Corrales Escobosa, Alma Rosa, Wrobel, Kazimierz
• Format: Article
• Language: English
• Subjects: Alkaline methanolysis; Antimony; Antimony - chemistry; Drinking Water - chemistry; Hydride generation; Least-Squares Analysis; Microwave plasma-atomic emission spectrometry; Microwaves; Partial least square regression; Polyethylene terephthalate; Polyethylene Terephthalates - chemistry; Spectrum Analysis
• ISSN: 0045-6535; 1879-1298
• DOI: 10.1016/j.chemosphere.2022.137316

Antimony is present in different types of plastics as a catalyzer residue and/or as a synergistic fire retardant; relatively high concentrations of this element reported in polyethylene terephthalate (PET) bottles and wrappers as well as its migration to the edible products or to different environment compartments are of concern. In this work, Sb determination is such products had been undertaken using hydride generation - microwave plasma - atomic emission spectrometry. To avoid harsh conditions typically reported for the digestion of PET, alkaline methanolysis was introduced whereas water samples were analyzed directly. Another original approach was to perform quantification by partial least squares regression (PLS1), taking spectral data from 2-nm range that comprised two emission lines (217.581 nm and less intense 217.919 nm). For PET, the calibration solutions contained Sb-free digest and covered the Sb concentration range 80–230 μg L−1. For the analysis of water, the calibration range was 0.5–10 μg L−1 and aqueous standard solutions were used. PLS1 provided reliable prediction, eliminating spectral interferences detected in the presence of PET digests and compensating for the spectral changes observed at low Sb concentrations. After standard addition to the real-world samples, the percentage recoveries were in the range 93.8–99.3% and 68–102% for PET and for bottled water, respectively. The method quantification limit for PET was 10 mg kg−1 and for water it corresponded to 0.20 μg L−1. The concentrations of Sb found in the analyzed samples were: 154–279 mg kg−1 for PET bottles and