No abstract

The conformation of lowest energy of an asphaltene molecule of the Athabasca sand oil was calculated through molecular mechanics. The molecule has a complex globular shape with small internal cavities. This shape resulted mostly from the existence of polymethylene bridges connecting the aromatic regions. Molecular aggregates formed with the asphaltene and with nine resins from the same oil, and with n-octane and toluene, were also studied. The resins showed higher affinities for the asphaltene than toluene and n-octane and also exhibited a noticeable selectivity for some of the external sites of the asphaltene. This selectivity based on the molecular recognition of the site depends on the fit between the resins and the site of the asphaltene. The selectivity explains why resins of one oil may not solubilize asphaltenes from other crudes. An analysis of the changes in the enthalpic and entropic contributions to the free energy showed that both contributions should be considered when the stability of the asphaltene and resin molecular aggregates is examined.

Problems surrounding molecular aggregation, covalent molecular weight, and their experimental investigation in asphaltene chemistry are reviewed. Chromatographic, fluorescence spectroscopic, and mass spectroscopic (MS) methods for the investigation of these problems are surveyed and their merits and limitations discussed. Dissociation of asphaltene in dilute solution can be followed in time by monitoring the gel permeation retention time variation with the age of the solution. This way, Athabasca asphaltene was reported to dissociate from several thousand to less than about 1000 g mol -1 molecular weight (MW) species in CH 2 Cl 2 solution to an extent of at least 80% in 14 days' time. The dissociation products represent the monomeric covalent molecules of asphaltene, and the remaining undissociated 20% could be slowly dissociating aggregates or high-MW covalent asphaltenes. The vapor pressure osmometry (VPO) determined number average MW of the same asphaltene was of the order of 4000 g mol -1 , manifesting the aggregated state of the asphaltene at the higher concentrations used for VPO measurements. Of the MS methods, the most thoroughly investigated and proposed to be the best suited to asphaltene studies are the laser desorption ionization/matrix-assisted laser desorption ionization (LDI, MALDI) timeof-flight (TOF) MS. However, results obtained from various laboratories do not compare well; in some cases the bulk of the m/z lies below 1000, and in others it lies well above m/z 1000. 252 Cf plasma desorption MS data are more self-consistent in the sense that the bulk of m/z values always lie below m/z 1000. The upper m/z limit in most cases is around a few thousand m/z but may extend up to tens of thousands m/z. The problems affecting these methods for the determination of covalent, monomeric asphaltene MW distributions are fragmentation of covalent bonds, multiple ionization, and the production of cluster ions. Fluorescence-based methods are not suitable for MW measurements in asphaltene; the reasons for this are discussed in detail.

Athabasca n-C 5 -asphaltene was fractionated into occluded maltene, low and high molar mass (LMA, HMA) asphaltene, and the latter fractions were subjected to Ni 2 B reduction to cleave the sulfide C-S bonds, basic hydrolysis to cleave the ester C-O bonds, and BBr 3 treatment to cleave the ether C-O bonds. Ni 2 B reduction of asphaltenes yielded 5-18% n-pentane solubles, which were separated into saturates, aromatics, and polars, and the saturates were analyzed for biomarkers. The residual asphaltene underwent 40% desulfurization and a greater than 4-fold drop in the MW of HMA but no change in the MW of LMA. The decrease in the MW is attributed to sulfide-bound core segments in the structure of the asphaltene:+ Ni 2 B f 4[core] + 3H 2 S. This is an important structural feature of Athabasca asphaltene and is responsible for its upgradability without excessive coke formation. The biomarkers of the asphaltene fractions were also characteristically different with regard to maturity status and composition. Both fractions yielded n-alkanes, cheilanthanes, regular steranes, hopanes, and gammacerane, and the LMA also contained dicyclic terpanes and C 21 -C 25 steranes. Noteworthy was the absence of diasteranes, which are the only steranes in the maltene. In terms of the 20S/(S + R) steranes and 22S/(S + R) hopanes parameters the maturity varies as maltene > LMA > HMA. This difference is a manifestation of the thermocatalytic nature of the maturation process and the protection of the macromolecular nature of the asphaltene against contact with external reagents. Ni 2 B reduction indicates that (1) the n-alkane products arise from n-alkyl substituted thiolane/thiane and thiophene and (2) C 27 -C 30 steranes are attached to the asphaltene by one S atom, and the C 21 -C 25 steranes and terpanes by two S atoms. Basic and BBr 3 hydrolyses of HMA showed that both ester and ether linkages of n-acids and n-alcohols are present and that the esters are of recent origin, whereas the ethers were derived from the original biotic source material of the bitumen.

After some general comments on the concept of asphaltene, outstanding problems relating to the molecular structure of Athabasca asphaltene are discussed in light of new results on aromaticattached appendages derived from ruthenium-ions-catalyzed oxidation (RICO). Detected were homologous series of R-branched C 1 -C 4 n-alkyl side chains up to C 30 -C 40 in an aggregate amount of ∼10% of the n-alkyl side chains, C 15 -C 20 regular isoprenoids, C 20 -C 28 cheilanthanes, C 27 -C 32 hopanes, C 27 -C 29 steranes, C 21 -C 24 pregnanes, and a number of branched hydrocarbons giving hydroxy carboxylic acids. The nature and distribution of these aromatic-attached biomarkers are similar but not identical to those reported to be attached to the asphaltene via a sulfide bridge. They may have originated from secondary biotic sources and became incorporated into the asphaltene via a Friedel-Crafts-type reaction. Additional, previously not considered reactions in the RICO of asphaltene are described, and aspects of the analytical procedures are reviewed. Also, a new protocol minimizing losses due to separations and volatility is discussed. Further structural elements of the asphaltene molecule were identified in the polar fraction of the asphaltene pyrolysis oil, including alkylpyridines and -quinolines, n-alkanoic/alkenoic acids, n-alkylamides (tentative), and n-alcohols. All straight-chain species were dominated by even carbon members. It is shown that contrary to recent erroneous suggestions in the literature, pericondensed aromatic units play a very minor role in the molecular structure of petroleum asphaltene.