This Ph.D. thesis describes an experimental and modeling investigation of the thermal conversion of coal and an experimental investigation of the emission of NO from char combustion in O2/N2 and O2/CO2 atmospheres. The motivation for the work has been the prospective use of the technology “Oxy-Fuel Combustion” as a mean of CO2 abatement in large scale energy conversion. Entrained Flow Reactor (EFR) experiments have been conducted in O2/N2 and O2/CO2 mixtures in the temperature interval 1173 K – 1673 K using inlet O2 concentrations between 5 – 28 vol. %. Bituminous coal has been used as fuel in all the experiments. Devolatilization experiments showed that the volatile weight loss was not affected by the change from N2 to CO2. Analysis by Scanning Electron Microscopy (SEM) and Brunauer-Emmett-Teller (BET) surface area of sampled char did not reveal differences between the two atmospheres either. Char conversion profiles, obtained from sampled char did not show differences in conversion rate between O2/N2 and O2/CO2 atmospheres across the span of O2 concentrations in the interval of reactor temperatures 1173 K – 1373 K. At the reactor temperatures 1573 K and 1673 K and an inlet O2 concentration of 5 vol. % it was found that char conversion rate was lowered in O2/CO2 compared to O2/N2. This is caused by the lower diffusion coefficient of O2 in CO2 (~ 22 %) that limits the reaction rate in zone III compared to combustion in O2/N2. Using char sampled in the EFR experiments ThermoGravimetric Analyzer (TGA) reactivity profiles for combustion and CO2 gasification has been found at 5 vol. % O2 or 80 vol. % CO2 and a heating rate of 5 K/min to a peak temperature of 1273 K or 1373 K. These experiments did not reveal differences in reactivity between EFR-chars formed in O2/N2 and O2/CO2. Reactivity profiles for TGA combustion of partly converted EFR-char, sampled at 1173 K and 28 vol. % O2, showed the presence of two phases of distinctively different reactivity. The least reactive of these phases are believed to be formed from interactions between mineral matter and secondary volatiles evolved during the fierce heating upon particle ignition at the high O2 concentrations. From TGA reactivity profiles of EFR-chars devolatilized at 1173 K – 1673 K intrinsic kinetic parameters has been found for combustion. The rate constant includes both a deactivation and an activation term. Intrinsic kinetics was also found for CO2 gasification though using only EFR-char devolatilized at 1273 K, 1473 K and 1573 K due to a lack of samples. Interestingly, it was found that devolatilization temperature did not affect the gasification rate constant. A detailed COal COmbustion MOdel (COCOMO) encompassing among others the three char morphologies; cenospheres, network- and dense chars, each distributed between six discrete particle sizes has been developed. The model showed a reasonable ability to predict the conversion profiles obtained in the EFR experiments using the intrinsic TGA kinetics for combustion and gasification. At the reactor temperature 1173 K COCOMO over predicts char conversion at O2 inlet concentrations of 5 and 28 vol. %. At the reactor temperature 1273 K COCOMO also over predicts char conversion at an inlet O2 concentration of 5 vol. %. Over prediction at the high O2 concentration is caused by the formation of char with a low reactivity as discussed above. Simulation under these conditions show that particle excess temperatures of 500 – 600 K is reached upon ignition at heating rates as high as 20000 K/s, which can indeed cause a significant release of secondary volatiles. At the low O2 concentration deviation between model and experiments is caused by an experimental delay in ignition that is not captured by the model. Though this causes a deviation in total conversion COCOMO still predicts conversion rates accurately after ignition. A laboratory scale Fixed Bed Reactor (FBR), operated isothermally at 1073 K, has been used for combustion of millimeter-sized lignite- and bituminous char particles in 5 – 80 vol. % O2 in N2 or CO2 atmospheres. Particle temperatures have been recorded by a Charged Coupled Device (CCD) camera and experiments have been carried out with single and multiple particles of different sizes. NO emission from lignite char were not affected by the change of N2 with CO2. Emissions from bituminous char were lower in O2/CO2. Emissions for both char types decreased as the O2 concentration or the particle size increased. An intermediate particle size was found where emissions peaked for the bituminous char. The CCD camera measured in situ temperatures accurately during experiments and the ability of film recording proved a valuable tool for data interpretation. The results suggest that transport phenomena and kinetics alone can not account for changes in NO emissions between O2/N2 and O2/CO2. The effect of mineral catalysis and the presence of other N-containing species, such as HCN and NH3, may also play a role.
Main Research Area:
Jensen, Peter Arendt, Jensen, Anker Degn
Technical University of Denmark, Department of Chemical and Biochemical Engineering, 2011