Modeling A Scaled 300 kWth Circulating Fluidized Bed Reactor With Barracuda

As global warming is continuing to gain recognition as a problem, new technologies for capturing greenhouse gasses such as CO₂ are being introduced.  The aim of these techniques is to capture CO₂ before it enters the atmosphere thus ameliorating its deleterious effects on global temperature and climate.  Chemical looping combustion with oxygen uncoupling (CLOU), a subset of chemical looping combustion (CLC), allows for isolation of carbon dioxide (CO₂) as an almost pure gaseous stream, suitable for compression and storage.  This is done by using two interconnected fluidized beds (CFB) with a metal oxygen carry capable of transferring oxygen from an air reactor to a fuel reactor.  The flue gas of the fuel reactor contains a highly concentrated stream of CO₂ and H₂O.  After condensing H₂O the CO₂ can be easily captured.  This eliminates any direct efficiency loss from the combustion process.  The goal of this study was to study fluid dynamics of a 300 kW_th CFB using computation particle fluid dynamic software Barracuda VR, then scaling this reactor to a cold flow model and demonstrating with Barracuda that similar fluid dynamics could be obtained.

The scaling was performed similar to Chalmers University of Technologies which was adapted from Levenspiel's text Fluidization Engineering.  This method uses dimensionless π-groups to determine the similarities of the beds.  Then from these an overall scaling value is determined and used to adjust the geometric dimensions of the CFB.  The time scale dimension is determined from raising the overall scaling value to the ½ power.  The particle density and fluid density ratio must be maintained for the cold and hot set up.  The hot system has a gas density of 0.236 kg/m³ and a particle density of 2140 kg/m³.  From these scaling parameters it was determined that helium (0.167 kg/m³) was the required fluidization gas and that a particle with density of 1450 kg/m³ would be required.  The particle of the hot system will fall within the Geldart A group and the particle in the cold system will fall in the Geldart A group.  Due to the price and difficulty of obtaining helium it was also determined that the scaled reactor would be run using air at 20 ^oC as the fluidizing gas.  The models to be meshed in Barracuda were created using Solidworks 2012.

Using Barracuda the following transient data was obtained and analyzed: oxygen carrier volume fraction (at points in the air reactor, at points in the air reactor riser, at point in the fuel reactor, and loop seals); the gas pressure at different points of the reactors.  The oxygen carrier mass was monitored at the outlet of the air reactor riser and the cyclone outlet.  The average residence time of the oxygen carrier was determined in the air reactor and fuel reactor.  The reactors were visual inspected using movies created in Barracuda.  These were compared for the hot reactor, cold reactor with helium as the fluidizing gas, and the cold reactor with air as the fluidizing gas.

From the Barracuda simulations it was determined that the hot and both cold reactor systems function with similar fluidization patterns.  It is recommended that the cold flow model should be constructed to study the effects of static electricity and other unknown inputs.  Then kinetics can be added to Barracuda to fully study the hot system.