DEVELOPMENT OF NANO-STRUCTURED MATERIALS THROUGH A NOVEL MULTI-SCALE MODELLING FRAMEWORK FOR ENERGY CONVERSION WITH CO 2 CAPTURE Shareq Mohd Nazir 1,* , Joana Francisco Morgado 1,2,5 , Stefan Andersson 3 , Zheng Xiao Guo 4 , Shahriar Amini 1,3 1 Norwegian University of Science and Technology, Trondheim, Norway 2 University of Coimbra, Coimbra, Portugal 3 SINTEF Industry, Trondheim, Norway 4 University College London, United Kingdom 5 Ifavidro Lda, Portugal Norwegian University of Science and Technology
Outline • Background and introduction • Different modeling scales – Atomistic level modeling – 1D modeling (reactor scale) – Process modeling (plant scale) • Description of the method – flow and type of data • Results • Summary Norwegian University of Science and Technology 2
Outline • Background and introduction • Different modeling scales – Atomistic level modeling – 1D modeling (reactor scale) – Process modeling (plant scale) • Description of the method – flow and type of data • Results • Summary Norwegian University of Science and Technology 3
Background Source: International Energy Agency (2017), Energy Technology Perspectives 2017, OECD/IEA, Paris Norwegian University of Science and Technology 4
CO 2 capture methods Post- combustion Power CO 2 Plant Capture Exhaust CO 2 Pre-combustion CO 2 CO 2 Fuel Capture CO 2 (coal, Gasification conditioning, natural Shift / Reforming H 2 transport and gas, oil, Syngas storage biomass) Power Plant CO 2 Power Water Plant removal Air separation Air unit Oxy-combustion O 2 Norwegian University of Science and Technology 5
Reforming methods Chemical Looping Steam Methane Partial Oxidation (POX) Reforming (SMR) Reforming (CLR) 𝐷𝐼 4 + 𝐼 2 𝑃 ⇌ 𝐷𝑃 + 𝐼 2 • CLR has less thermodynamic losses and has inherent air SMR separation 𝐷𝐼 4 + 𝑃 2 ⇌ 𝐷𝑃 + 𝐼 2 POX • CLR reforms CH 4 to a product gas with higher H 2 /CO ratio 𝐷𝐼 4 + 𝑁𝑓𝑃 ⇌ 𝐷𝑃 + 𝐼 2 CLR when compared to conventional POX Norwegian University of Science and Technology 6
Earlier project Project NanoSim : A Multiscale Simulation-Based Design Platform for Cost-Effective CO 2 Capture Processes using Nano-Structured Materials (EU FP7 framework) https://www.sintef.no/projectweb/nanosim/ 1. System Scale • Develop an open-source computational platform that will 2. Equipment Scale allow the rational design of the second generation of 3. Cluster Scale gas-particle CO 2 capture technologies based on nano- 4. Particle Scale structured materials 5. Intra-particle pore scale • Design and manufacture nano-structured material and 6. Atomistic scale shorten the development process of nano-enabled Consortium products based on the multi-scale modelling 1. SINTEF Industry 2. TU Graz • Design and demonstrate an energy conversion reactor 3. University College London with CO 2 capture based on the superior performance of 4. INPT Toulouse nano-structured materials 5. NTNU 6. DCS Computing GmbH 7. Andritz Energy and Environment GmbH 8. University de Coimbra Norwegian University of Science and Technology 7
Outline • Background and introduction • Different modeling scales – Atomistic level modeling – 1D modeling (reactor scale) – Process modeling (plant scale) • Description of the method – flow and type of data • Results • Summary Norwegian University of Science and Technology 8
Atomistic and cluster scale modeling • Reactivity of nanoparticles at the atomic scale/nanoscale, is estimated through kinetic Monte Carlo (kMC) modeling, guided by Density Functional Theory (DFT) calculations, on the detailed kinetics of the CH 4 conversion to products as a function of temperature. • Cluster scale: o Intra-particle transport model o Fluid-Particle flow model (Tools: LIGGGHTS for particle motion and CFDEM for fluid flow) Reference: Andersson, S., et al., Towards rigorous multiscale flow models of nanoparticle reactivity in chemical looping applications. Catalysis Today, 2019. Norwegian University of Science and Technology 9
Equipment scale - 1D Model of CLR • Rapid convergence • Wide range of applicability (reasonably generic) • User friendly • Accommodate reactor clusters • Handle dynamic and stationary simulations “Generalized fluidized bed reactor” (GFBR) model Bubbling Turbulent Fast fluidization Norwegian University of Science and Technology 10
1-D model for fluidized bed reactors Generic formulation based on the generic model developed by Abba et al. (2003) • uses an averaging probabilistic approach Single formulation is used! • Two-phase model Differential Balances Numerical scheme: • Mass balance • Method of lines (MATLAB routine ode15s ) • Gas total mass balance • Finite volume method (discretization in • Gas species mass balance for space) each phase • Non-uniform grid • Total solids species mass • Convective term : 1 st order upwind balance scheme • Diffusion term : central differences • Total Energy balance scheme • Pressure Balance Reference: Abba, I.a., et al., Spanning the flow regimes: Generic fluidized-bed reactor model. AIChE Journal, 2003. 49 : p. 1838-1848. Norwegian University of Science and Technology 11
Two phases Syngas (CO+H 2 ) + Depleted air CO 2 + H 2 O 𝑈 𝑡,𝐵𝑆 𝑥 𝑡,𝐵𝑆 𝐻 𝑡,𝐵𝑆 𝐿 MeO Me L-Phase H-Phase Air Fuel ψ 𝐼 , 𝑣 𝐼 , ψ 𝑀 , 𝑣 𝑀 , reactor reactor 𝜁 𝐼 , 𝐸 ,𝐼 𝜁 𝑀 , 𝐸 ,𝑀 𝑈 𝑡,𝐺𝑆 𝑥 𝑡,𝐺𝑆 𝐻 𝑡,𝐺𝑆 Air CH 4 + Sketch of the two-phase H 2 O approach for a fluidized bed Simulation domain reactor Norwegian University of Science and Technology 12
Averaging probabilistic approach • Library of closures for different fluidization regimes Fast Bubbling Turbulent Fluidization Regime Regime Regime j=1 j=2 j=3 𝜄 2 𝜄 1 𝜄 3 = 𝑸 𝟐 𝜄 1 + 𝑸 𝟑 𝜄 2 + 𝑸 𝟒 𝜄 3 𝜄 𝑸 𝒌 Probability of being under regime j Reference: Abba, I.a., et al., AIChE Journal, 2003. 49 : p. 1838-1848. Norwegian University of Science and Technology 13
1D Model outline Constants Initial and Boundary conditions Reactor dimensions; Fundamental and kinetic Thermochemical properties cosnstants Relations for gas and solids properties Differential Balances Reaction kinetics Mass balance; Energy balance; Pressure Closures for hydrodynamics Bubbling, Turbulent and Fast Fluidization Regimes Solver Probabilistic Approach Define the model hydrodynamic Simulation results parameters Norwegian University of Science and Technology 14
Parameter interaction in 1D-Model KMC – Kinetic Monte Carlo • Kinetic parameters (Arrehnius parameters) Affects: Gas physical properties/conditions Gas velocities • Flowrate Void fraction • Density Temperature • Composition Reaction rate (R = k C n ) • Heat capacity Pressure drop Solid physical properties/conditions Affects: • Flowrate Solids velocities • Density Void fraction • Composition Temperature • Reaction rate (R = k C n ) Temperature • Heat capacity Pressure drop • Particle size Solid recirculation rate Norwegian University of Science and Technology 15
System (process plant) scale model Thermodynamics Hydrodynamics Thermodynamics + Kinetics MATLAB Norwegian University of Science and Technology 16
Interaction between 1d model and plant scale simulations Norwegian University of Science and Technology 17
Pre-combustion combined cycle with CLR (CLR-CC) Norwegian University of Science and Technology 18
Key performance indicators 𝐷𝑃 2 𝐷𝑏𝑞𝑢𝑣𝑠𝑓𝑒 = 𝐷𝑃 2 𝑓𝑜𝑓𝑠𝑏𝑢𝑓𝑒 𝑗𝑜 𝑢𝑖𝑓 𝑞𝑠𝑝𝑑𝑓𝑡𝑡 × 100 CO 2 Capture (%) 𝐷𝑃 2 𝑓𝑛𝑗𝑢𝑢𝑓𝑒 𝑐𝑧 𝑠𝑓𝑔. 𝑞𝑚𝑏𝑜𝑢 −𝐷𝑃 2 (𝑓𝑛𝑗𝑢𝑢𝑓𝑒 𝑐𝑧 𝑞𝑠𝑝𝑑𝑓𝑡𝑡) = × 100 CO 2 Avoidance (%) 𝐷𝑃 2 (𝑓𝑛𝑗𝑢𝑢𝑓𝑒 𝑐𝑧 𝑠𝑓𝑔. 𝑞𝑚𝑏𝑜𝑢) 𝑂𝑓𝑢 𝐹𝑚𝑓𝑑𝑢𝑠𝑗𝑑𝑗𝑢𝑧 𝑄𝑠𝑝𝑒𝑣𝑑𝑓𝑒 = 𝑀𝐼𝑊 𝑝𝑔 𝑔𝑣𝑓𝑚 𝑗𝑜𝑝𝑣𝑢 𝑢𝑝 𝑞𝑠𝑝𝑑𝑓𝑡𝑡 × 100 Net Electrical Efficiency (%-LHV input) Norwegian University of Science and Technology 19
Key performance indicators 𝑀𝐷𝑃𝐹 = (𝑈𝐷𝑆)(𝐺𝐷𝐺) + 𝐺𝑃𝑁 Levelised cost of electricity ( 𝑁𝑋 )( 𝐷𝐺 × 8766) + 𝑊𝑃𝑁 + ( 𝐼𝑆 )( 𝐺𝐷 ) ($/MWh) TCR – Total capital requirement FOM – Fixed operating & maintenance costs FC – Fuel costs VOM – Variable operating & maintenance costs HR – Heat Rate $ 𝑀𝐷𝑃𝐹𝐷𝑀𝑆 − 𝐷𝐷 − 𝑀𝐷𝑃𝐹𝑂𝐻𝐷𝐷 𝐷𝑝𝑡𝑢 𝑝𝑔 𝐷𝑃 𝑏𝑤𝑝𝑗𝑒𝑓𝑒 ) = Cost of CO 2 avoided 𝑢𝐷𝑃 2 𝑁𝑋ℎ 𝑂𝐻𝐷𝐷 − 𝑢𝐷𝑃 2 𝑢𝐷𝑃 2 𝑁𝑋ℎ 𝐷𝑀𝑆 − 𝐷𝐷 *GCCSI. 2013. Global CCS Institute - TOWARD A COMMON METHOD OF COST ESTIMATION FOR CO 2 CAPTURE AND STORAGE AT FOSSIL FUEL POWER PLANTS. Norwegian University of Science and Technology 20
Recommend
More recommend
