Boiling point modeling for petroleum speciation data

Abstract

An empirical model is presented for generating the boiling curve of a petroleum sample based on parameters that are experimentally available using typical speciation techniques. The approach estimates the boiling point distribution of the detected components and allows a comparison with the simulated distillation profile of the sample. The model is shown to predict the boiling behavior of the speciation-derived average components in crude oil distillation fractions, which match the measured average distillation temperature of the fractions. The model is then applied to generate boiling distributions, based on mass spectrometric and gas chromatographic speciation data, which match the measured simulated-distillation curves with less than 20 °C difference. The model is also applied to the isolated fraction of saturated compounds, where the modeled boiling profile matches well with the measured distillation profile of the entire sample. This confirms that the model, although built using Fourier transform ion cyclotron resonance mass spectrometry data of aromatic compounds, extends to data obtained using comprehensive 2-dimensional gas chromatography, and also describes the boiling behavior of saturated compounds speciated by field desorption time of flight mass spectrometry. Consequently, the modeled boiling distribution enables an assessment of how well a speciation measurement describes the sample, and provides an estimation of lower and higher-boiling portions that have been omitted in the data set.