Document e7DdggGZQNvL1dLd3V9ZLERxm

EFFECT OF PRESSURE ON DELAYED COKING OF ATHABASCA BITUMEN Jagannathan Govindhakannan and Chandra Khulbe Natural Resources Canada, 1 Oil Patch Drive, Devon, AB Canada Introduction Delayed coking, the most widely used coking process, was developed to minimize refinery yields of residual fuel oil by severe thermal cracking of feedstocks such as vacuum residue and thermal tars (Gary and Handwerk, 1984). The delayed coking process is particularly attractive when the green coke produced can be sold for anode or graphitic carbon manufacture or when there is no market for fuel oils, as in western Canada (Gray, 1994). The process uses long retention times in the liquid phases to convert the residue fraction of the feed to gases, distillates, and coke. The thermal cracking reactions produce olefinic species in the liquid product and the condensation reactions produce highly aromatic coke product that retains sulfur, nitrogen, and metals. The effects of process conditions such as pressure, temperature, and residence time are important in determining the yield, selectivity, and quality of delayed coking products. Although plenty of information is available in the open literature regarding the effects of temperature and residence time on delayed coking products, very few data are presented on the effect of pressure. In the present work, we summarize the results of in-house pilot plant work carried out using Athabasca bitumen to study the effect of pressure on the yield and quality of delayed coking products. Experimental A schematic diagram of the delayed coker pilot plant is shown in Figure 1. The feedstock is loaded into a 200-L feed tank, which is typically heated to about 100C to allow pumping of heavy oil/bitumen. The feed is then pumped through preheaters to the bottom of one of two coker drums. Typical feed rates range from 1 to 5 kg/h. Water supplied by a pulseless pumping system is fed into a heated coil and vaporized. The steam joins the oil at the entrance of the oil preheater, in which the steam and feedstock are heated further to the coker inlet temperature. The coker drum has a replaceable sleeve-liner, which is inserted to protect the inside of the coker drum and the end-cap seals. The liner also facilitates removal of the coke from the drum after the run. The sleeve is replaced after each run. In the coker drum, the oil cracks into lighter liquids, noncondensible gases, and coke. The vapors, noncondensible gases, and steam leave the coker drum at the top and flow to a hot separator that is typically kept at 300C. The liquid level in this vessel is controlled by adjusting the flow rate of the heavy ends (HE). HE are collected in a receiver maintained at atmospheric pressure. The remaining vapors/gases and steam are cooled and collected in a cold separator. From the cold separator, a light ends (LE) stream is removed that contains water and is collected in a receiver at atmospheric pressure. Uncondensed gases, together with solution gases from the light- and heavy-end receivers, flow through a pressure controller, which regulates the system pressure. They then pass through a dry test meter, and finally to a vent system. A slip stream of all gases, including process gas and dissolved gas, is sent to an on-line gas chromatograph for detailed compositional analysis. At the end of a run the feed is stopped while steam continues to flow. After this steaming period, the drum heaters in use are turned off and the drum is cooled. After the drum and coke have been cooled the coke deposits are collected and stored. Results and discussion A full-range Athabasca bitumen was used as the feedstock in the delayed coker pilot plant. The chemical and physical properties are given in Table 1. All four runs were conducted under identical conditions except for the coker drum pressure, which was varied from 20 psig to 150 psig. The effects of coker drum pressure on the yields of total liquid product (TLP), gas, and coke are shown in Figure 2. Pressure increase suppresses vaporization in the coker drum and consequently increases the residence time of the liquid, thus allowing more cracking to occur. Due to the increased cracking, the yields of gas and coke increase and the yield of TLP decreases. The chemical and physical properties of TLP at different operating pressures are presented in Table 2. As the operating pressure increases, sulfur and nitrogen conversion increase as indicated by their reduced concentrations in the TLP. The simulated distillation data with respect to different operating pressures indicate that the TLP becomes progressively lighter as the operating pressure increases. It must be noted that the yield of naphtha (IBP177C) fraction increases with increasing pressure similar to that of the light gas oil (177260C) fraction. The increases in the yields of the naphtha and light gas oil fractions were offset by the decrease in the yield of the residue fraction (316C+). Chemical and physical properties of the naphtha, light gas oil, gas oil, and residue fractions obtained with respect to different operating pressures were analyzed. The coke properties, presented in Table 3, indicate that the concentration of sulfur and nitrogen decreases in the coke as the operating pressure increases, which is inline with the observation that the conversion of sulfur and nitrogen increases as the operating pressure increases. Conclusions Four delayed coking runs were completed successfully at different coker drum pressures while keeping other process conditions identical. The results from the experiments indicate that the yield of TLP decreases and the yields of coke and gases increase as the operating pressure increases. Sulfur and nitrogen conversions increase with increasing pressure. Acknowledgements Partial funding for this work has been provided by the Canadian Program for Energy Research and Development (PERD), the Alberta Research Council (ARC), and the Alberta Energy Research Institute (AERI). References Gary, J. H., Handwerk, G. E. Petroleum Refining, Second edition, Marcel Dekker, Inc. New York. 56, 1984. Gray, M. R. Upgrading petroleum residues and heavy oils, Marcel Dekker, Inc., New York. 244, 1994. Prepr. Pap.-Am. Chem. Soc., Div. Petr. Chem. 2010, 55 (1), 44 Table 1 Chemical and Physical Properties Bitumen of Athabasca Analysis Units Value Carbon wt% 84.32 Hydrogen wt% 10.43 Nitrogen wt% 0.39 Aluminum mg/kg 6.2 Barium mg/kg 3 Calcium mg/kg 26.1 Copper mg/kg 1.9 Iron mg/kg 13.4 Magnesium mg/kg 1.6 Manganese mg/kg 0.2 Molybdenum mg/kg 9.6 Sodium mg/kg 41.9 Nickel mg/kg 80.3 Silicon mg/kg 862 Titanium mg/kg 2.4 Vanadium mg/kg 218 Arsenic mg/kg 0.2 Potassium mg/kg 0.7 Sulfur wt% 4.67 Density g/mL 1.008 API gravity API 8.88 ASH wt% 0.04 Heptane insoluble wt% 11.37 Hexane insoluble wt% 11.2 Pentane insoluble wt% 15.03 Toluene insoluble wt% 0.01 Microcarbon residue wt% 13.108 Molecular wt g/gmol 525 Dynamic viscosity 60C cp 2810 Dynamic viscosity 100C cp 187 Table 2 Chemical and Physical Properties of TLP at Different Operating Pressures Analysis IBP177C 177260C 260316C 316C+ Density at 15.6C API gravity Carbon Hydrogen Nitrogen Sulfur Ash Pentane insolubles Heptane insolubles Hexane insolubles Toluene insolubles Microcarbon residue Molecular wt Viscosity at 25C Viscosity at 40C Units wt% wt% wt% wt% g/mL API wt% wt% mg/kg wt% wt% wt% wt% wt% wt% wt% g/mol cSt cSt 20 psig 11.78 10.17 14.38 63.66 0.9149 23.01 84.84 11.4 1666.9 3.28 0.017 0.058 0.009 0.041 0.026 0.751 395 9.471 5.993 40 psig 15.99 15.23 15.98 52.80 0.9013 25.34 85 11.74 1502.75 3.18 0.002 0.05 0.033 0.014 0.011 0.5 427 5.587 3.9 150 psig 25.51 24.21 16.43 33.84 0.8716 30.69 85.08 11.79 1086.23 2.77 0.001 0.054 0.018 0.03 0.036 0.457 466 2.464 1.874 Table 3 Coke Properties at Different Operating Pressures Analysis Density at 15.6C Carbon Hydrogen Nitrogen Sulfur Pentane insolubles Toluene insolubles Calorific value Aluminum Calcium Copper Iron Magnesium Manganese Molybdenum Sodium Nickel Silicon Titanium Vanadium Arsenic Mercury Moisture Ash Volatile matter Unit g/mL wt% wt% wt% wt% wt% wt% Btu/lb mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg mg/kg ng/g wt% wt% wt% 20 psig 1.389 85.62 3.97 1.74 6.45 99.464 97.445 14651 1.4 21.5 1.8 19 1.4 0.5 15 42.4 214 10.4 4.2 575 <0.1 24 0.21 1.72 9.08 40 psig 1.383 88.84 3.75 1.66 6.36 99.769 97.321 14915 2.1 35.8 20 66.5 2 1.5 35.2 147 352 33.4 7.7 926 0.2 38 0.3 0.43 10.65 150 psig 1.347 85.7 5.67 1.2 5.96 80.743 65.946 14854 10.6 50.8 103 242 3.7 4 23.5 79.7 283 34.1 5 672 0.2 34 2.66 7.67 30.16 HOT BITUMEN FEED CABINET COKE DR UMS FEED TANK M NI TROGEN WATER HOT SEP LT LV COLD SEP LT LV WT WT RECEIVERS WT WT RECEIVER S PROCESS VENT PROCESS VENT Figure 1 Process flow diagram of delayed coker pilot plant Yield, Wt% 90 80 TLP Coke Gas 70 60 50 40 30 20 10 0 0 20 40 60 80 100 120 140 Pressure, psig Figure 2 Effect of pressure on TLP, coke, and gas yields 160 Prepr. Pap.-Am. Chem. Soc., Div. Petr. Chem. 2010, 55 (1), 45