A simple, fast and green method for API gravity, density, sulfur and nitrogen determination in crude oil by ATR-FTIR

Abstract

In this work, the combination of multivariate data analysis and Fourier-transformed infrared (FTIR) for the characterization of crude oil samples based on the determination of API gravity (°API), density, sulfur and nitrogen contents is proposed. Reference values for density and °API were determined according to ASTM D7042-04, while nitrogen was determined according to ASTM D4629. For sulfur determination, a miniaturized MAD-SRC method was used to obtain reference values. The best model for density and °API was obtained applying a partial least squares with two response variables (PLS2) calibration. Multiplicative scattering correction (MSC) and mean centering pre-treatments were applied for this model (R2 of 0.936 and 0.944 for density and °API, respectively). For N content, the best PLS model consisted in using selected variables and first derivative with Savitsky-Golay smoothing (R2 = 0.983). The PLS model for S content prediction was built using selected variables and applying first derivative with Savitsky-Golay smoothing and mean-centering data pre-treatment (R2 = 0.982). The environmental impact and greenness of crude oil characterization applying both the alternative (using multivariate calibration) and the official methods was evaluated using the White Analytical Chemistry (WAC) tool. WAC analysis resulted in relatively low scores for conventional methods (76.8 for ASTM D4629, 75.8 for ASTM D7042-04 and 70.7 for MAD-SRC) while the ATR-FTIR method was considered the greener method (low impact, WAC score of 98.3). The proposed method combining FTIR with multivariate calibration enabled the use of a very low sample volume (about 0.01 mL), minimum waste generation, and suitable results were achieved, saving time and reagents.