DELAYED COKING OF SOLVENT EXTRACTED COAL FOR PRODUCTION OF ANODE GRADE COKE: CHARACTERIZATION OF SOLID AND LIQUID PRODUCTS

Open Access
- Author:
- Karri, Vamsi
- Graduate Program:
- Energy and Mineral Engineering
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- None
- Committee Members:
- Caroline Clifford Burgess, Thesis Advisor/Co-Advisor
Caroline Elaine Clifford, Thesis Advisor/Co-Advisor - Keywords:
- solvent extraction
coprocessing
delayed coking
anode grade
GC-MS
co-coking - Abstract:
- This study investigates the feasibility of using high temperature solvent extraction of coal to produce feedstock for the production of anode grade coke through delayed coking. Solvent extraction runs were performed with Western Kentucky No. 6 coal and Conoco Philips decant oil at a solvent to coal ratio of 10:1, 390 oC and 200 psi (1.37 MPa) N2 pressure in a 2 L stirred autoclave followed by hot filtration at 200 oC. The conversions varied from 8% to 57% (dry basis) based on mass using ash as tracer, due to issues with the filter. Multiple runs were performed in order to generate enough material for the laboratory scale delayed coking experiments. The low conversions were attributed to the high H/C ratio of the decant oil and the reactor design, particularly the filtration unit of the reactor. Coal extraction, under conditions used in this work, proceeds through the thermal relaxation which enhances the transport properties and thermolysis, possibly, to crack enough of the coal structure in order to release the compounds of coal into the extract. Characterization of the decant oil and the coal extract was performed using simulated distillation GC and GC-MS. The coal extraction resulted in a decrease in the single- and two-ring aromatic compounds concentration and increased the concentration of the polyaromatic hydrocarbons. The concentration of the paraffins also increased slightly. Even though there was increase in the concentration of two ring compounds in the jet fuel range (boiling point 180 oC – 270 oC), there was an overall increase in the concentration of the polyaromatic hydrocarbons because the fuel oil fraction (boiling point >330 oC) was the largest fraction in the decant oil. Cokes were produced by delayed coking in a laboratory-scale delayed coker at the EMS Energy Institute. Decant Oil was also coked for comparison purposes. A total of three coking reactions were performed. The solvent extracted liquid was coked at 470 oC and 0.273 MPa (39.7 psig), and the decant oil was coked at 465 oC and 0.253 MPa (36.8 psig). A catalytic coking run with the solvent extracted liquids was also performed at 467 oC and 0.263 MPa (38.2 psig). The catalyst used in the process is one that is typically used in fluid catalytic crackers, although no specific information about the catalyst was provided. Higher pressures (40 psig) compared to typical refinery conditions (10-25 psig) were employed to increase the yield of coke. The gasoline fraction obtained from coking of the decant oil and the coal extract was mostly paraffinic while the catalytic coking of the coal extract had a higher concentration of single-ring aromatic compounds indicating that the catalyst used aided in the formation of aromatic compounds in the liquid fractions. The jet fuel fraction also showed similar trends as the gasoline fraction. The heavier fuel fractions had an increase in the concentration of single, two-ring aromatic compounds and polyaromatic hydrocarbons while the concentration of the paraffin decreased. The extensive cracking during the delayed coking process resulted in the conversion of the paraffins into aromatic compounds (liquid and solid aromatic compounds) and light gases. It was observed that the catalyst increased the liquid yields and decreased the solid coke yield, and may have increased the rate of hydrocarbon cracking in the coker. The cokes produced had low ash contents ranging from 0.54-1.01% by mass. Calcination and graphitization of the cokes at 1400 oC and 3000 oC reduced the ash contents drastically, with the ash content of the graphitized cokes less than 0.15% by mass. The cokes produced were characterized using Optical Microscopy, X-ray diffraction (XRD), Raman spectroscopy and Temperature Programmed Oxidation (TPO). Optical microscopic analysis of the cokes from coal extract (OTI - 14.8) were slightly less anisotropic than the coke from decant oil (OTI - 20) due to the introduction of coal derived structural elements. The green cokes generated had similar lattice parameters (interlayer spacing ~ 3.420 Å, Lc ~ 27 Å) as determined by XRD indicating that the chemical composition of the feed was not altered enough to affect the graphitability. Raman spectroscopic analysis showed that the calcined cokes had a high degree of disorder (~ 45%) which decreased to less than 10% upon graphitization. TPO analysis showed that the green cokes were composed of filamentous carbon which on calcination and graphitization transformed into crystalline graphite platelets. Comparison of the liquids and cokes generated through co-coking of coal with decant oil showed that the solvent extraction process produces cokes of higher quality as indicated by the optical texture index (14.87 for coal extract compared to 7.1 for co-cokes), lattice parameters and ash contents (0.57% by mass for coal extract compared to 1.02% by mass for co-cokes) while the co-coking produces liquids that could potentially be more valuable (higher concentration of two-ring aromatics in liquids from co-coking) depending on the product composition and the end use of the liquids.