| Texto completo | |
| Autor(es): |
de Moraes-Neto, Venancio Ferreira
;
Baqueta, Michel Rocha
;
Carames, Elem Tamirys dos Santos
;
Rocha, Rodrigo Barros
;
Pallone, Juliana Azevedo Lima
Número total de Autores: 5
|
| Tipo de documento: | Artigo Científico |
| Fonte: | Food Chemistry; v. 493, p. 13-pg., 2025-07-30. |
| Resumo | |
Robusta Amazonico coffee is a highly valued specialty coffee with a registered geographical indication, making it a potential target for fraud. In this context, near-infrared (NIR) and ultraviolet-visible (UV-Vis) spectroscopy are highlighted as environmentally friendly analytical techniques with strong potential for verifying authenticity. In this study, 114 samples of Robusta Amazonico green coffee and 108 samples of other coffees (in intact, ground, and extracted forms) were analyzed using both benchtop and handheld NIR instruments, as well as a UV-Vis spectrophotometer. The Data-Driven Soft Independent Modeling of Class Analogy (DD-SIMCA) algorithm was employed for classification, and performance was evaluated based on sensitivity, specificity, and accuracy. Classification accuracy ranged from 95.6 % to 100 %. Notably, the handheld NIR achieved 100 % accuracy and proved to be the most cost-effective tool, highlighting its suitability for verifying the authenticity of Robusta Amazonico green coffee. (AU) | |
| Processo FAPESP: | 22/03268-3 - Técnicas analíticas alternativas e ensaios de digestão in vitro para alimentos especiais |
| Beneficiário: | Juliana Azevedo Lima Pallone |
| Modalidade de apoio: | Auxílio à Pesquisa - Regular |