Portal de Periódicos da CAPES

Artigo Acesso aberto Produção Nacional Revisado por pares

BIBTEX RIS

An inexpensive NIR LED Webcam photometer for detection of adulterations in hydrated ethyl alcohol fuel

2017; Elsevier BV; Volume: 135; Linguagem: Inglês

10.1016/j.microc.2017.08.014

1095-9149

Hebertty V. Dantas, Mayara Ferreira Barbosa, Anderson Lopes Pereira, Márcio José Coelho Pontes, Pablo Nogueira Teles Moreira, Mário César Ugulino de Araújo,

Spectroscopy Techniques in Biomedical and Chemical Research

In this study an inexpensive near infrared light emitting diodes (NIR LEDs) photometer using a webcam as a transmittance detector is proposed for the first time. As an illustrative application, the proposed NIR LED Webcam (NLW) photometer was used to detect adulterations of hydrated ethyl alcohol fuel (HEAF) by water or methanol. The proposed photometer used seven NIR LEDs in peak wavelengths of 805, 845, 902, 912, 930, 1173 and 1282 nm as its radiation source, and a common CMOS (Complementary Metal-Oxide-Semiconductor) webcam, which was connected to the USB port of a notebook as its detector. One hundred and forty-seven HEAF samples, being 100 adulterated (with water 52 samples, and methanol 48 samples), and 47 unadulterated samples were used to evaluate the performance of the proposed NIR LED Webcam (NLW) photometer. Digital images acquired by webcam were decomposed in histograms of color systems, RGB, Grayscale, HSV, HLS, CMYK using a publicly available graphical interface. The histograms were then submitted to multivariate classifiers: the successive projections algorithm for variable selection in association with linear discriminant analysis (SPA-LDA) and partial least squares discriminant analysis (PLS-DA). For both SPA-LDA and PLS-DA models, all samples in the test set were correctly classified using the RGB, HSV, CMYK, and Grayscale histograms. The proposed NLW photometer using chemometric tools was efficient in detecting unadulterated and adulterated HEAF samples.