Some heavy oil and bitumen production processes involve the precipitation of asphaltenes upon the addition of a paraffinic diluent. For example, in a new oil sands process, a paraffin solvent is used to treat the froth generated in bitumen extraction. The recently proposed VAPEX process is also based on a paraffinic solvent. To optimize these processes, data is required on asphaltene precipitation from heavy oils or bitumens diluted with n-alkanes.
In this work, the onset and amount of asphaltene precipitation from Athabasca bitumen were investigated in a PVT cell using three different methods over a range of temperatures and pressures. Results are presented for the precipitation of asphaltenes from the bitumen diluted with nalkanes, including n-heptane, n-pentane, n-butane, and propane.
Chemicals and Materials
Two batches of Athabasca coker feed bitumen were obtained from Syncrude Canada Ltd. This oil sand bitumen had been processed to remove water and solids, such as sand and clays, and was ready for upgrading. Table 1 shows the SARA analysis of the two samples. n-Heptane and npentane were obtained from Aldrich Chemical Company and were 99%+ pure. n-butane and propane were obtained from Praxair and were 99.5% pure.
PVT Experiments
A mercury-free PVT cell was used to study the effect of pressure and temperature on asphaltene precipitation for bitumen-alkane systems. The cell capacity is approximately 100 cm3 and a typical sample size was 35 cm3. The precipitation experiments in the PVT cell were performed using three different methods: the "premixing method", the "heating method", and the "bench-top method".
The development and testing of the premixing method is described in detail elsewhere 1,2 and is outlined below. The bitumen was premixed with solvent at a low solvent-to bitumen ratio to reduce viscosity sufficiently to flow the bitumen. The premixed solvent-to-bitumen ratios were 0.51, 0.32, and 0.24 wt/wt for n-heptane, n-pentane, and n-butane, respectively. The bitumen and solvent were premixed in a sample cylinder, sonicated, and shaken for 2 hours for the nheptane- bitumen mixture and for 10 hours for the other bitumen-solvent mixtures. The final solvent-to-bitumen ratio was determined from the mass of solvent in the bitumen plus the mass of solvent initially charged into the cell.
Note that a potential source of error is that the premixed bitumen/solvent mixture may not be homogeneous. If the mixture is not homogeneous, the solvent-to-bitumen ratio of the sample displaced into the PVT cell may not be the same as the ratio in the sample cylinder.
Before charging the PVT cell, it was first evacuated using a vacuum pump for 10–15 minutes. The PVT cell was then filled with the required volume of solvent. The connecting lines were drained and evacuated once again and the required volume of premixed solvent and bitumen was added to the cell. The contents of the cell were mixed continuously with a magnetic stirrer for 23 hours. The sample was then displaced from the PVT cell to a modified sample bomb.
