Chemical Looping Combustion Cold Flow Model commissioning and performance evaluation
MetadataVis full innførsel
- Institutt for elkraftteknikk 
SINTEF and NTNU are planning to build a 150 kWth Chemical Looping Combustion (CLC) reactor system. This is new technology and the CLC reactor system is going to be one of the largest of its kind in the world. The technology is promising for CO2 capture in terms of energy efficiency and economics. To verify the design a Cold Flow Model, CFM, has been built. In the CFM no reactions take place, but it simulates the hydrodynamics of the 150 kWth CLC reactor system. The reactor system consists of two reactors exchanging solids in a loop. The two reactors are one air reactor, AR, and one fuel reactor, FR. Air is injected at different locations in the CFM to fluidize the solids and achieve the proper mass flows. The Cold Flow Model has been commissioned and an experimental campaign was executed. A series of experiments running each reactor singularly were performed. The rig seems to be functioning satisfactory and a minimum of plugging in the pipes were observed. The Cold Flow model has two cyclones that showed collection efficiencies at approximately 99 %. This is important to avoid emissions of solids from the future CLC reactor system, both for economic and environmental reasons. An investigation and mapping of the operating area of the reactors singularly and coupled was the target of the experiments. Correlations between operating velocity, total solid inventory, air distribution and flux were found. Appropriate flow regimes, meant to give good gas solid contact efficiency, and mass flow’s entrainments were achieved. The targets of a solid circulation rate of 2 kg/s in the AR and 1 kg/s in the FR were also achieved. Air is injected in the bottom of the reactors to fluidize the particles. This air is distributed through primary and secondary nozzles. The highest primary air percentage tested in the FR, 75%, gave the highest flux. In the AR 100% was tested, but 70% gave the highest flux. The last result is in contradiction with other experimental work in the area which says that 100% primary air should give the highest flux. After the mapping of the operating area of the single reactors it was possible to try to run the two reactors coupled. The divided loop seal was tested but led to a pressure short circuit and a large amount of the total solid inventory was lost out of the cyclones in a short time. The operation of a divided loop seal is probably possible, but seems difficult. The internal part of the loop seals were sealed to make the operation easier. The loop seals could then be operated as traditional loop seals. A challenge was the mass balance between the fuel reactor and air reactor. The mass flows of particles from both reactors must be equal to have a mass balance. Otherwise all the particles eventually ends up in one reactor. Results from the single reactor experiments were used to know approximately which operating conditions gave a mass balance between the reactors. The Cold Flow Model seemed to a certain degree be self regulating for achieving a mass balance if initial operating conditions were reasonable. Two experiments with coupled reactors and mass exchange only through the loop seals were done. A global solid circulation rate of 0.7 kg/s and 1 kg/s was achieved. Both AR and FR had the proper flow regimes. Proper flow regimes in the reactors are turbulent or fast fluidization. A third experiment utilized a lifter to enhance the solid transport between the reactors. A lifter is a additional transporter of solids from one reactor to another. The lifter worked successfully. The experiment had a global solid circulation rate of 1.4 kg/s. The mass flows were 1.4 kg/s from the AR loop seal and 1 kg/s from the FR loop seal. The remaining part 0.4 kg/s from the FR to the AR was transported with the lifter. Both reactors had proper flow regimes. A fourth experiment trying to achieve a global solid circulation rate of 2 kg/s failed. The bottleneck seems to be the AR loop seal. Solids accumulated and the loop seal was not able to handle this rate of solid flow. A new operation philosophy and design of the loop seal has been proposed. The new design of the loop seal and operation philosophy reduces the air flow needed in the loop seal, but it may not necessarily solve the solid circulation limit in the AR loop seal. Further investigation is needed. Manipulating the pressure in the AR may contribute to enhance the rate of solid flow through the loop seal. The successful experiments were presented at the 1st International Conference on Chemical Looping, IFP-Lyon, France, 17 - 19 March 2010. After the experimental campaign was finished the experiments were simulated with the fluidization software ERGUN developed by Compiegne University of Technology. ERGUN applies different mathematical models. For the simulations performed Horio’s and Berruti’s model were applied. The evaluation of the ERGUN simulations by means of the experiments shows that Horio’s and Berruti’s model should not be used for a detailed investigation of the flow structure in the CFM’s risers. However, despite its strongly empirical nature, a preliminary investigation of the riser’s behavior with Berruti’s model may be useful. Berruti’s model is a reasonable tool for modeling the upper part of the pressure profile in the AR and FR at the operating conditions tested. The operating conditions tested in the AR are total solid inventories of 35 and 45 kg, and superficial gas velocities from 0.9-1.9 m/s. The operating conditions tested in the FR are total solid inventories of 35 and 50 kg, and superficial gas velocities from 1.5-2.0 m/s. Berruti’s model is not capable of accounting for the dense bed in the lower part of the reactor as Horio’s model does. However, Horio’s model mismatched the experimental results too much. Horio’s model seems to be a provide a better match at larger total solid inventory and smaller operating velocities, hence flow regimes not relevant for the CLC reactor system.