Rapid Purity Determination of Copaiba Oils by a Portable NIR Spectrometer and PLSR

Copaiba oil is a non-timber forest product utilized in popular medicine, in addition to the pharmaceutical, cosmetic, and food industries. Brazil is a chief producer and supplier of this oil, and due to an increasing demand and elevated market value, it causes for this product to be subject to adult...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Food analytical methods 2018-07, Vol.11 (7), p.1867-1877
Hauptverfasser: de Oliveira Moreira, Alessandro Cézar, de Lira Machado, Angelo Henrique, de Almeida, Fernanda Vasconcelos, Braga, Jez Willian Batista
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Copaiba oil is a non-timber forest product utilized in popular medicine, in addition to the pharmaceutical, cosmetic, and food industries. Brazil is a chief producer and supplier of this oil, and due to an increasing demand and elevated market value, it causes for this product to be subject to adulterations. Thus, the goal of this study was to develop a direct, fast, and simple method to quantify the purity of the oil extracted from Copaifera langsdorffii Desf., specifically, when the oil was adulterated with soybean oil. Quantification was performed utilizing a portable NIR spectrometer and partial least squares regression (PLSR). In the development and validation of the method, 53 samples of copaiba oil expressing a purity ranging from 50 up to 100% were used. Of the 53 samples, 15 were pure, 31 were adulterated with unused soybean oil, 6 were adulterated with soybean oil used for frying, and 1 adulterated with an unknown vegetable oil. Four models were developed and the best among them presented RMSEP = 1.5%, R 2  = 0.991, and REP lower than 2.0% and expressed precision with deviations below 0.7%. These results indicate that the method is suitable for quality control analysis. In addition, it was accurate in the identification of samples with that were not present in the developed model. Graphical Abstract ᅟ
ISSN:1936-9751
1936-976X
DOI:10.1007/s12161-017-1079-8