Coal direct chemical looping process: 250 kW pilot-scale testing for power generation and carbon capture

Application of an equation‐oriented framework to formulate and estimate parameters of chemical looping reaction models

“…Inspection of the experimental OC conversion vs. time plots of CH 4 in Figure 5 for these runs shows that the model fits begin to worsen with a reduction in temperature and the model does not capture the inflection behavior of the experimental curves which generally begins at OC conversion greater than 40% at which the cracking of gaseous reactants on the metallic Fe sites most likely begins to become significant. These deviations do not significantly affect the ability of the model to be predictive in our chemical looping applications of interest in the areas of combustion and reforming, because in these applications a high circulation rate of the OC between the reducer reactor and oxidation reaction is maintained to keep the temperature in the reducer between 950 and 1050°C for thermal management reasons 41 . At these high circulation rates, reduction of the OC in the reducer does not go above 40% 41 (after which point the deviations and inflection of the experimental curves are observed for CH 4 ), and thus the model's application is restricted to the region where the fit is good.…”

“…These deviations do not significantly affect the ability of the model to be predictive in our chemical looping applications of interest in the areas of combustion and reforming, because in these applications a high circulation rate of the OC between the reducer reactor and oxidation reaction is maintained to keep the temperature in the reducer between 950 and 1050 C for thermal management reasons. 41 At these high circulation rates, reduction of the OC in the reducer does not go above 40% Further investigation of these exceptions shows that the CI values of k 0,2 (AE193%) and k 0,3 (AE141%) for CH 4 in Table 8 have large variations from the mean values indicating the estimates are illconditioned with respect to errors in the reactant gas concentration and highlighting the potential to improve these estimates with additional data and alternative model forms that better capture concentration dependence for reactions (R8) and (R9). In addition, it can be observed from Table 9 that k 0,3 and n 3 also have wide CI values (AE73:4% and AE78:1%, respectively) suggesting that there might be a need for more experimental data on the reduction of the OC with CO to better determine these parameters (reduce the uncertainty) and better characterize the Fe 2 O 3 to Fe 3 O 4 reduction step for reaction (R4).…”

“…In our study, , provided by Equations () and (), is the relationship between reactant gas concentration, , and the input measured variables (). and are measured by MFCs with relative standard deviations of , 41 while the pressure controller and temperature controller have relative standard deviations of and respectively 42 . Based on this, the maximum relative standard deviation in the reactant gas concentration for all the reactant gases and across all experimental runs was found to be .…”

“…41 (after which point the deviations and inflection of the experimental curves are observed for T A B L E 8…”

Coal-Direct Chemical Looping Process with In Situ Sulfur Capture for Energy Generation Using Ca–Cu Oxygen Carriers

“…The CDCL process has been successfully demonstrated at various scales ranging from a 2.5 kW th bench unit to a 250 kW th pilot unit. A 25 kW th sub-pilot scale unit designed at commissioned at OSU has been subjected to 200 h of continuous operation, whereas the 250 kW th pilot unit has been operated for 288 h for power generation and CO 2 capture with high coal conversion and CO 2 purity. − …”