RESUMO
INTRODUCTION: Red fruit oil (RFO) is a natural product extracted from Pandanus conoideus Lam. fruit, a native plant from Papua, Indonesia. Recent studies indicate that RFO is popularly consumed as herbal medicine. Therefore, the quality of RFO must be assured. OBJECTIVES: This study aimed to develop a chemometric analysis applied to 1 H nuclear magnetic resonance (NMR) and Fourier transform infrared (FTIR) data for important quality parameter distinction of red fruit oil (RFO), especially regarding the degree of unsaturation and the amount of free fatty acids (FFA). MATERIALS AND METHODS: Forty samples consisting of one crude RFO, thirty-three commercial RFOs, and three oils as blends, including olive oil, virgin coconut oil, and black seed oil, were analysed by 1 H NMR and FTIR spectroscopy. After appropriate preprocessing of the spectra, principal component analysis (PCA) and partial least squares regression (PLSR) were used for model development. RESULTS: The essential signals for modelling the degree of unsaturation are the signal at δ = 5.37-5.27 ppm (1 H NMR) and the band at 3000-3020 cm-1 (FTIR). The FFA profile represents the signal at δ = 2.37-2.20 ppm (1 H NMR) and the band at 1680-1780 cm-1 (FTIR). PCA allows the visualisation grouping on both methods with > 98% total principal component (PC) for the degree of unsaturation and > 88% total PC for FFA values. In addition, the PLSR model provides an acceptable coefficient of determination (R2 ) and errors in calibration, prediction, and cross-validation. CONCLUSION: Chemometric analysis applied to 1 H NMR and FTIR spectra of RFO successfully grouped and predicted product quality based on the degree of unsaturation and FFA value categories.