ORNL is managed by UT-Battelle, LLC for the US Department of Energy COMSOL Model for Optimizing Regeneration of CFP Catalyst in Packed-Bed Reactors COMSOL Meeting 2019 Bruce Adkins and James Parks II Oak Ridge National Laboratory Kristiina Iisa and Kristin Smith National Renewable Energy Laboratory www.cpcbiomass.org
16
Embed
COMSOL Model for Optimizing Regeneration of CFP Catalyst ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
ORNL is managed by UT-Battelle, LLC
for the US Department of Energy
COMSOL Model for Optimizing Regeneration of CFP Catalyst in Packed-Bed Reactors
COMSOL Meeting 2019
Bruce Adkins and James Parks IIOak Ridge National Laboratory
Kristiina Iisa and Kristin SmithNational Renewable Energy Laboratory
Catalyst Load Testing at National Energy Technology Lab (NETL)
21
%
Range of DensitiesCompares Very Well to Commercial Sock vs Dense Loading
Methods
77 www.cpcbiomass.org
Model Deliverables1. Regeneration
– Temperature distribution during coke combustion is critical information
• Large rapid swings in temperature can physically damage catalysts
➢ Particle confinement stress
𝝈𝒑𝒄𝒔 = 𝑬𝒃𝒎𝜶𝒕𝒄∆𝑻
• Ebm = bulk modulus of compressibility, Pa
• atc = thermal expansion coefficient, K-1
➢ Variations in heating rates
• High temperatures can accelerate deactivation (Ostwald ripening)
2. Upgrading
• Flow (mal)distribution
• Catalytic performance
• Deactivation by coke & contaminants
3. Full cycle
• Optimize individual steps and overall cycle
• Predict long term performance
• Scale-up to commercial scale
CURRENT
FOCUS
88
COMSOL Model Elements
Coke Type Ea, J/mol Af, s-1
Soft 60,964 598
Medium 72,680 223
Hard 130,798 71,870
ChemistryModule
Coke Combustion
C+O2→CO2DH = 3.94x105 J/mol
H2+½O2→H2ODH = 2.40x105 J/mol
N2 as Solvent
Transport of Dilute Species Module
Surface Species OR NondiffusiveVolume Species
Reactive Pellet Bed
Reactions
Intraparticle DiffusionDp = ? m2/s
Bed PropertiesParticle Shape & Dimensions
Bed Porosity = ?
Film Resistance?
Bulk DiffusionD = ? m2/s
Laminar FlowModule
Porous BedKozeney-Carman
Permeability
Free ConvectionFurnace Cavities
Bulk ConvectionN2 & Air Properties
Heat TransferModule
Particle ThermalProperties
Discretization of Xtra Dim.
Reactor ThermalProperties
Heat Source
N2 & Air ThermalProperties
Heat Release
Particle PropertiesSurface Area = ? m2/gPore Volume = ? cm3/g
Skeletal Density = ? kg/m3
Therm Conduct = ? W/m.KHeat Capacity = ? J/kg.K
Geometry
Full 3D OR2D Axisymmetric
Mesh
Kinetic Constants for Petroleum Cokes
Ea = ? J/molAf = ? m3/mol.s
Solvers
Time Depend. BDF
Rel. Tolerance?
Segregated vs. Fully Coupled
Direct vs. Iterative
99
Different Versions for Different Needs
2D Axisymmetric ModelsAlso 3D Models
21 Versions to Date
Bed Slice: 1-5 g CatalystDiscretization Studies
Flo
w
Full Bed: 100 g CatalystInitial Validation Studies
Flo
w
Complete SystemFull Validation Studies
Inlet
CatalystBed
FurnaceCavity Reactor
Cavity
FurnaceWalls
UpperT-coupleTC-223
LowerT-coupleTC-265
ReactorBody
Outlet
Flo
w
1010 www.cpcbiomass.org
Discretization of Extra Dimension in Reactive Pellet Bed Model
Catalyst A
Ea = 5x104 J/mol
Af = 5x104 m3/mol.s
Dp = 1x10-5 m2/s
1,050 cells lavg = 1.32 mm
Bed Slice
Flo
w
atxd2(r,z,tds.rpb1.cpes_cC_surf)
1111 www.cpcbiomass.org
Discretization of Extra Dimension in Reactive Pellet Bed Model
Af = 2x105 m3/mol.s
Dp = 1x10-6 m2/s
Catalyst A
Ea = 5x104 J/mol
Af = 5x104 m3/mol.s
Dp = 1x10-5 m2/s
1212 www.cpcbiomass.org
Partial Validation with Full Bed Model
TE-265
Flo
w
Catalyst A
Ea = 5x104 J/mol
Af = 1x105 m3/mol.s
Dp = 1x10-5 m2/s
1313 www.cpcbiomass.org
CompleteSystemModel
Inlet
CatalystBed
FurnaceCavity Reactor
Cavity
FurnaceWalls
UpperT-coupleTC-223
LowerT-coupleTC-265
ReactorBody
Outlet
Temperature°C
1414 www.cpcbiomass.org
Cumulative Particle Confinement Stress – One Regeneration Cycle
𝜎𝑝𝑐𝑠 = න0
𝑡𝑟𝑒𝑔𝑒𝑛
𝐸𝑏𝑚 ∗ 𝛼𝑡𝑐 ∗ 𝑇 − 𝑇𝑖𝑛𝑖𝑡 𝜕𝑡
Anatase TiO2
Ebm = 300 x 109 Pa
atc = 12 x 10-6 K-1
spcs
Pa.s
1515 www.cpcbiomass.org
Summary
1. COMSOL has been used to build a computational model of a packed bed CFP reactor in regeneration mode. A key ingredient is the Reactive Pellet Bed (RPB)
2. Coke combustion in this system exhibits classical “cigar burn” behavior, and the kinetic model is a stiff system of equations. Non-physical results, mass closure errors and rapid divergence can result from excessively coarse discretization
• It is important to thoroughly test the effects of:
➢ discretization in the pellet extra dimension in RPB, and
➢meshing in the bed
for each combination of:
➢ intrapellet diffusion rate
➢ reaction rate
➢ pellet geometry, and
➢ bed geometry
• Time stepping relative tolerance is also important to numerical stability. Typical value is 0.001
3. The model is still being tuned at 100 gram scale with cylindrical catalyst
4. Extensions of the model to 0.5 mm spherical catalyst and 10+ kg scale are in progress
1616 www.cpcbiomass.org
Acknowledgements
• DOE EERE Bioenergy Technology Office (BETO)
• Jeremy Leong, Sponsor
• ChemCatBio consortium
• Bryan Hughes and Tim Floyd, NETL
This work was authored in part by the National Renewable Energy Laboratory, operated by Alliance for Sustainable Energy, LLC, for the U.S. Department of Energy (DOE) under Contract No. DE-AC36-08GO28308. Funding provided by the U.S. Department of Energy Office of Energy Efficiency and Renewable Energy Bioenergy Technologies Office. The views expressed in the article do not necessarily represent the views of the DOE or the U.S. Government. The U.S. Government retains and the publisher, by accepting the article for publication, acknowledges that the U.S. Government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this work, or allow others to do so, for U.S. Government purposes.