Evaluation of metal content in tea samples commercialized in sachets using multivariate data analysis techniques

Tea is a beverage consumed all over the world, and, besides the very pleasant taste, it has substances in its composition that can lead to various beneficial effects to human health. However, although teas have essential elements in their composition, they can also be contaminated with metals from t...

Full description

Saved in:

Bibliographic Details
Published in:	Microchemical journal Vol. 151; p. 104248
Main Authors:	Gomes, Délis Alves Souza, Alves, Juscelia Pereira dos Santos, da Silva, Erik Galvão Paranhos, Novaes, Cleber Galvão, Silva, Darci Santos, Aguiar, Rosane Moura, Araújo, Sulene Alves, dos Santos, Ana Caroline Lessa, Bezerra, Marcos Almeida
Format:	Journal Article
Language:	English
Published:	Elsevier B.V 01-12-2019
Subjects:	Hierarchical cluster analysis Kohonen maps Metals Neural networks Principal component analysis Teas in sachets Teas in sachets Kohonen maps Neural networks Metals Hierarchical cluster analysis Principal component analysis
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Tea is a beverage consumed all over the world, and, besides the very pleasant taste, it has substances in its composition that can lead to various beneficial effects to human health. However, although teas have essential elements in their composition, they can also be contaminated with metals from the soil, air, and equipment used in their production. In this study, eight metals (Ca, Cu, Fe, Mg, Mn, Zn, Na and K) were determined in tea samples commercialized in sachets using flame atomic spectrometry (absorption and emission) after acid decomposition in a digestion block. The concentration of the analyzed metals varied as follows (in mg kg−1): Ca (1856.3–10,012), Cu (2.014–14.90), Fe (43.94–532.2), Mg (739.9–2237), Mn (26.95–946.3), Zn (12.05–41.84), Na (167.7–4322) and K (5089.1–14,334). The generated data were statistically analyzed using the following multivariate analysis tools: Principal Component Analysis, Hierarchical Cluster Analysis and Kohonen self-organizing maps. The multivariate analysis classified the different tea samples into well defined groups according to flavor, based on the mineral composition of eight quantified elements. •Metals were determined in tea samples commercialized in sachet using F AAS after acid digestion;•Mutivariate analysis techniques were employed to evidence latent information;•Principal component analysis, hierarchical cluster analysis and artificial neural network associated with Kohonen maps were compared;•Tea samples show similarities in relation the kind of plant used as raw material.
ISSN:	0026-265X 1095-9149
DOI:	10.1016/j.microc.2019.104248