(546e) Understanding Distillation Curves and Pseudocomponents

Refineries process streams with thousands of components. Refinery streams defeat conventional analytic methods for defining precise chemical compositions of the feeds, intermediates, and products. Conventional approaches use pseudo-components to characterize refinery streams. While long established, few plant engineers understand the assumptions and limitations of pseudo-components.

Data for pseudo-component generation comes from distillation curves generated in specific ASTM methods. In rare cases, non-ASTM methods, can also generate distillation data. Distillation methods are well established. However, plant engineers often do not understand their limitations.

The combined lack of knowledge on both distillation curves and pseudo-component generation often leads to useless and deceptive simulations. Three case studies present common problems with refinery simulations. These include:

• Crude distillation
• Hydrotreating
• Catalytic reforming

Particular focus is placed on the following issues:

• How distillation curves are interpreted, converted and used
• Data accuracy on distillation curves and how that introduces modeling errors
• Data accuracy on sampling and how data that does not mass balance creates errors in simulations

These problems consistently undermine design, operating, and troubleshooting efforts with models. Understanding them, and how to reduce their effect, is critical to effective modeling for plant operations.