Rapid Analytical Characterization Of Residues From Heavy Oil And Bitumen Upgrading Processes

Abstract

The upgrading of heavy oil and bitumen usually produces some high boiling residues. In order to evaluate these materials so that they may be effectively utilized, component type analyses are useful. One of the standard methods for the analysis of component types (SARA) consists of the precipitation of asphaltenes followed by chromatographic separation of the remainder of the sample into saturates, aromatics and resins fractions using columns of silica gel, alumina and clay. This procedure is long and tedious and not that suitable for routine screening of large numbers of samples. The latroscan method (thin layer chromatography with flame ionization direction) provides a component type analysis which is faster and more convenient than SARA. The total elapsed time for the analysis of a set of ten samples simultaneously by the former method is less than two hours compared to about two days per samples by the SARA procedure. A series of residue samples from different upgrading processes using Cold Lake, Lloydminster and Athabasca feed-stocks was examined using the latroscan method. Based on the wide range observed in the quality of the samples, which in some cases could be correlated with the feed-stock and the type of upgrading process, suggestions may be made for possible uses of the residues. Introduction The processing of heavy oil and biltumen usually results in some high boiling residue or pitch material. Effective utilization of these residues requires their analytical characterization by such methods as softening point, viscosity, density, carbon residue, ash and elemental analysis. Often a classification according to component type is necessary particularly if it is suspected that the residue may be of a quality suitable for further upgrading into useful products. This type or analysis usually requires an initial determination of the asphaltene content by precipitation with a low molecular weight paraffin such as n-pentane followed by chromatographic separation of the deasphalted portion of the sample using column of silica gel, alumina and clay. This separation usually yields such fractions as saturates, aromatics and resins, and is commonly referred to as the SARA analysis(1). Unfortunately this procedure requires about two days or elapsed time per sample, is long and tedious and therefore not suitable for routine screening of large numbers of samples or for the monitoring of products from a process. A method employing thin layer chromatography with flame ionization detection (TLC/FID) using an latroscan analyzer does provide a rapid and convenient component type analysis of heavy hydrocarbons. Several workers have already applied techniques based on the latroscan to the characterization of bitumens heavy oils and coal liquids(2-7). The latroscan method, as it has been developed in our laboratory has been described previously(8). In the present work it is shown how the method was used to characterize residues from a variety of upgrading processe employing Cold Lake, Lloydminster and Athabasca feed-stocks. This characterization was based upon sample separation into saturates, aromatics and polar fractions to produce semi-quantitative data. Preliminary results are also given to show how the aromatics may be further separated to provide additional characterization.

Hydrocarbon wastes at petroleum- and creosote-contaminated sites: rapid characterization of component classes by thin-layer chromatography with flame ionization detection

“…This procedure has been employed widely in the petroleum industry for the component class characterization of heavy oils, bitumens, and coal-tar liquids (9-15). Fuhr et al (9,12,13) described application of the technique for the characterization of bitumens from oil sand deposits and for process monitoring during the upgrading of heavy oil residues. Poirier et al (15) used TLC-FID for the class analysis of residues obtained from the coprocessing of bitumen and sub-bituminous coal in order to establish the degree of coal conversion in the liquefaction process.…”

Estimation of the SARA Composition of Crude Oils from Bubblepoint Pressure Data

“…Figures – show respectively the aromatic SARA compositions versus the saturate SARA compositions; the resin SARA compositions versus the saturate SARA compositions; the asphaltene SARA compositions versus the saturate SARA compositions, and the aromatic SARA compositions versus resin SARA compositions. In all cases, the data were taken from our database, − …”

“…In this work, we would rather use one W for each oil type (bitumen/heavy-oil/medium-oil/light-oil). For the set of 341 crude oil compositions studied in this work, − …”

“…The probability density W , for an oil, was estimated by the histogram of SARA compositions taken from the data set. − As occur in all histograms in statistics, if the mesh size is big enough, it ...…”

“…Table shows some features of our database. The set includes SARA analyses reported in 108 public papers found in open and private literature. ,− …”