The purpose of this study was to create a theoretical model for the distillation of synthetic crude oil (syncrude) into straight-run naphtha, kerosene and gas oil fractions, at a scale of 235 ktfeed/year. To that end, a process model was created that receives the output of a Co-LTFT process as raw feed. The raw feed was prefractionated to remove the majority of the unreacted syngas, inert gases, light hydrocarbons (C1-C4) and water. The remaining stream was syncrude, primarily composed of alkanes and 1-alkenes in the C5-50 range.
A Base Case was designed in ASPEN plus, with a distillation unit with two steam strippers and three pumparounds. The goal was to receive the syncrude and separate naphtha, kerosene and gas oil cuts, with least 90% purity and recovery of the components. The Fenske-Underwood-Gilliland method was used to estimate the number of stages and reflux ratio of the column, and the Kirkbride method was used to find an initial estimate for the feed stage. The TBP curves of the cuts were compared with similar cuts from literature and found to be similar.
The Base Case underwent a sensitivity analysis in order to ascertain the effect of different design and process parameters on the separation quality and utility consumption. The parameters that were tested are cold condensate temperature, feed stage, condenser duty, feed temperature and stripping steam flow rate to the ADU. According to the sensitivity analysis results, for optimal separation between the syncrude and the other gases in the raw feed, the gases must be purged at a temperature of -70°C. Furthermore, the optimal configuration for the distillation of the given syncrude into naphtha, kerosene and gas oil fractions with at least 90% purity and recovery is as follows: The ADU has 40 equilibrium stages and a condenser duty of approximately -5.7 MW. The feed must be heated to 310°C and enter the column at stage 37. The stripping steam flow rate must be around 1.3 kg/s.
Five alternative processes were modelled as well, with similar inputs to the column model, and the results of the distillate separation quality and utility consumption were documented as well. The alterations of the alternative cases include replacing the stripping steam with a reboiler in one of the strippers, adding a stripper from where an additional product was drawn, concentrating the pumparound duty on the condenser, using a vacuum distillation unit to fractionate the residue of the atmospheric distillation unit, and employing a heat integration network.
The most important conclusions are listed. Firstly, the side strippers must use low pressure stripping steam instead of reboilers. Next, the minimum number of products must be drawn off in order to minimize utility consumption at a given separation quality. Furthermore, heat integration can save up to 50% in total utility consumption (heat and cooling water). It is recommended that further research, including a cost estimation, is conducted on refineries that produce on-specification final products.