This Open File Report makes available additional geochemical and other data to supplement that reported by the authors in two papers, with the same general title, published in the February, 1983 issue of Geochimica Cosmochimica Acta.
With respect to the starting materials, differences between duplicate analyses of trace elements in the bitumen emphasize the tremendous difficulty in cleaning bitumen prior to analysis. A complete grain size distribution for the shale is provided without further comment.
For sixteen experiments in the complete set of 96, the bitumen was separated into four fractions (saturates, aromatics, resins and asphaltenes) following the hydrothermal treatment. Although there are differences in composition between the subsets shale and no-shale, as well as between the sixteen bitumens and the starting bitumen, their magnitude is insufficient to indicate that the experimental conditions were severe enough to induce major changes in the composition of the bitumen.
The composition of the colloidal material found in most of the experiments with shale has been calculated in terms of oxides, and for many of the twelve analyses reported, the Al:Si ratios fall in a fairly narrow range, even though the total amount of colloidal material present varies widely. This is interpreted in terms of a common mineral component in suspension in these cases.
An annotated ANOVA is presented in the Appendix for the individual element concentrations for the experiments with shale.
Of the 24 experiments which effectively represent 'blank' runs to test contamination effects of the autoclaves, only those eight of 92 hours duration were carried out, some in triplicate to test autoclaves of different materials. The chemical composition of the aqueous phase is given for sixteen of these 'blank' runs and, with the exception of Cr, it can be stated that contamination from the autoclaves is effectively negligible.
The final set of data reported are the aqueous phase compositions from the 24 experiments designed to test the effects of pH, temperature, time and salinity on the removal of trace elements from bitumen. The most dramatic effect is the reduction in pH to the range 2.5 to 3.2 for those experiments in which pH was not controlled by the addition of borax. Increased sulphate at higher pH values may be due to hydrolytic desulphurization reaction of the sulphonic acid group of the bitumen.