Font Size: a A A

Correlation and modeling of thermal cracking yields in the delayed coking process

Posted on:2005-12-14Degree:Ph.DType:Dissertation
University:The University of TulsaCandidate:Almerri, Hamad AliFull Text:PDF
GTID:1451390008989234Subject:Engineering
Abstract/Summary:PDF Full Text Request
With the increasing production of crude oils from reservoirs to meet market demand, the barrel of the crude oil becomes heavier with an increase of sulfuric compounds and heterogeneous atoms. Fortunately, these problematic components are successfully eliminated or reduced through the delayed coking process to avoid poisoning the catalyst in any downstream catalytic cracker or hydrotreater unit. In addition, the delayed coker can, through thermal cracking, convert 50--70% of the heavy vacuum resid into valuable liquid products ranging from gasoline to gas oil. Nevertheless, relatively little research has been done on delayed coking, especially about its thermal cracking behavior. Therefore, an extensive experimental program was carried out to investigate and model thermal cracking reactions in delayed coking.;To study the thermal cracking reactions, 37 experimental runs were performed in a stirred-batch reactor for six different resids where the gas, liquid, and coke were analyzed through elemental analysis and simulated distillation. During the experiments, resid samples were drawn from the reactor at different temperatures ranging from 775°F--850°F. Additionally, over 150 thermogravimetric tests (TGA) were conducted for the six resids and the resid samples drawn from the reactor. The stirred-batch and TGA results were used to develop a kinetic model with three parallel reactions that produce gasoline, diesel, gas oil, gas, and an intermediate that forms coke. The model uses one activation energy and three frequency factors for the three reactions to successfully predict the production of liquid and its sub-products within +/- 2.8% error of the experimental data.;In addition to the kinetic model, correlations were developed to predict product yields in the stirred-batch reactor, using feedstock properties such as API, MCR, H, H/C, C7 asphaltenes, and C5 asphaltenes. Finally, a correlation to predict the true boiling point of the vacuum resids was developed through feed properties and TGA.
Keywords/Search Tags:Thermal cracking, Delayed coking, Oil, TGA, Model
PDF Full Text Request
Related items