Hamid Arastoopour Linden Professor of Engineering and Director of - PowerPoint PPT Presentation

Hamid Arastoopour Linden Professor of Engineering and Director of Wanger Institute for Sustainable Energy Research (WISER) Illinois Institute of Technology, Chicago, IL

 Prof. Hamid Arastoopour (PI)  Prof. Javad Abbasian (Co ‐ PI)  Emad Abbasi (PhD Candidate)  Shahin Zarghami (PhD Candidate)  Emad Ghadirian (PhD Student)  Jaya Singh (PhD Student)

The overall objective of the program is to develop a Computational Fluid Dynamic (CFD) model and to perform CFD simulations to describe the heterogeneous gas-solid absorption and regeneration and WGS reactions in the context of multiphase CFD for a regenerative magnesium oxide-based (MgO-based) process for simultaneous removal of CO 2 and enhancement of H 2 production in coal gasification processes. 3

The Project consists of the following four (4) tasks: Task1. Development of a CFD/PBE model accounting for the particle (sorbent) porosity distribution and of a numerical technique to solve the CFD/PBE model. Task2. Determination of the key parameters of the absorption and regeneration and WGS reactions. Task3. CFD simulations of the regenerative carbon dioxide removal process. Task4. Development of preliminary base case design for scale up

Conventional Integrated Fuel Gas Fuel Gas T: 350°-500° C P: 10-70 bar WGS WGS Reaction Reaction CO 2 & CO 2 Removal CO 2 CO 2 Removal Concentrated Hydrogen Stream Chemical Syn./ Liquid Fuels Concentrated Hydrogen Stream Impurities Hydrogen Polishing Fuel Cells, Transportation

Concentrated CO 2 Hydrogen Stream Clean WGS Reaction XO Coal Gas Sorbent & make up Regeneration CO 2 Removal CO + H 2 O CO 2 + H 2 XCO 3 XO + CO 2 XO + CO 2 XCO 3

Sorbent Preparation, Characteristics and Reactivity

14 0.5 M K2CO3 CO 2 Capacity, g of co2/100 g of sorbent 0.7 M K2CO3 12 1 M K2CO3 2 M K2CO3 10 8 6 4 2 0 0 0.1 0.2 0.3 0.4 0.5 K/Mg Molar ratio 8

Preparation Parameters HD52 ‐ P2 Sorbent particle diameter,  m 150 ‐ 180 Calcination temperature,  C 520 Calcination temperature ramp,  C/min 1 Duration of calcination, hr 8 Concentration of potassium carbonate in the impregnation solution, mol/lit (M) 1 Duration of impregnation, hr 20 Drying temperature,  C (post ‐ impregnation) 90 Humidity during drying, % ambient Duration of drying, hr 24 Re ‐ calcination temperature,  C (post ‐ drying) 470 Calcination temperature ramp,  C/min 1 Duration of re ‐ calcination, hr 4 9

P P MFC CO 2 T P P MFC N 2 Vent Bubble Flow meter P Pressure Regulator T Data Acquisition & Control System 10

50% New sorbent 40% MgO Conversion, % 30% Old sorbent 20% 10% 0% 0 10 20 30 40 50 Time, min 11

14 12 CO 2 Capacity. Gco2/100g of sorbent 450 ˚ C 10 425 ˚ C 490 ˚ C 8 390 ˚ C 6 4 2 340 ˚ C 0 0 5 10 15 20 25 30 35 40 45 50 time, min 12

60% 50% MgO Conversion, % 40% 30% Initial Gas Composition 20% CO 2 /N 2 /H 2 O: 50/20/30 %mol* CO 2 /N 2 /H 2 O: 50/40/10 %mol* CO 2 /N 2 /H 2 O: 50/45/5 %mol* CO 2 /N 2 /H 2 O: 50/50/0 %mol 10% P=20 bar T=420 ˚ C 0% 0 5 10 15 20 Time,min *The sorbent is exposed to steam for 30 min prior to the run. 13

 Structural changes  Secondary Carbonation Reaction  MgO + H 2 O = Mg(OH) 2 Hydration  Mg(OH) 2 + CO 2 = MgCO 3 + H 2 O Carbonation 14

100 Hydration MgO+H 2 O Mg(OH) 2 Absorption 10 PCO2 & PH2O , bar MgO+CO 2 MgCO 3 1 Decomposition Dehydration MgCO 3 MgO+CO 2 Mg(OH) 2 MgO+H 2 O 0.1 0.01 175 225 275 325 375 425 475 525 575 Temperature, ˚ C 15

60% 50% MgO Conversion, % 40% 30% 20% Gas Composition CO 2 /N 2 /H 2 O: 50/20/30 %mol CO 2 /N 2 /H 2 O: 50/50/0 %mol 10% P=20 bar Reaction Time= 5 min 0% 280 300 320 340 360 380 400 420 440 460 Temperature, C MgO+CO 2 = MgCO 3 Mg(OH) 2 +CO 2 = MgCO 3 +H 2 O 16

1.E-03 k 1 Gas Film r p, k 1 1.E-04 k, (cm/min) Product Layer 1.E-05 Highly Reactive Zone (k 1 ) Low Reactive Zone (k 2 ) 1.E-06 k 2 r c k 2 1.E-07 r p ’ 0 100 200 Radius of Particle ( μ m)     X     D D ( ) r r ( 1 X ) ZX 3 g g 0 p p  k for r r   M 1 c  k   product react Z s  k for r r  M 2 c react product 2 k 3   s ( C C )( 1 X ) 3 b e o r N dX p   MgO dt 1  k 1 X    s 1 r ( 1 X ) ( 1 ) 3 3 p   Scale: D 1 X XZ 20 µm Abbasi et al., Fuel, 2013 g A. Hassanzadeh, 2007

Coupled Computational Fluid Dynamics (CFD) Population Balance Model (PBM ) (CFD ‐ PBM)

Numerical Modeling: Conservation Equations Eulerian- Eulerian Approach in combination with the kinetic theory of granular flow Assumptions: Uniform and constant particle size and density ‐ Conservation of Mass          ‐ gas phase: ( ) .( v ) m g g g g g g  t   ‐ solid phase        ( ) .( v ) m s s s s s s  t ‐ Conservation of Momentum  ‐ gas phase:                    ( v ) .( v v ) P . g ( v v ) g g g g g g g g g g g gs g s  t  ‐ solid phase                      ( v ) .( v v ) P P . g ( v v )  s s s s s s s s s s s s gs g s t ‐ Conservation of Momentum  ‐ gas phase:        ( y ) .( v y ) R  g g i g g g i j t  ‐ solid phase        ( y ) .( v y ) R  s s i s s s i j t ‐ Conservation of solid phase fluctuating Energy  3                     [ ( ) .( ) v ] ( p I ) : v .( ) ‐ solid phase s s s s s s s s s s  2 t Generation of Diffusion dissipation energy due to solid stress tensor Abbasi and Arastoopour , CFB10, 2011

Numerical Modeling: Drag Correlation Gas-solid inter-phase exchange coefficient: EMMS model ( Wang et al. 2004)    ( 1 ) 3 g g     Heterogeneity Factor   u u C ( ) 0 . 74 g g s D 0 g g 4 d ω < 1 p       ( 1 ) u u   2  ( 1 ) sg   0 . 74 g g s g g g  150 1 . 75 g  2 d d p g p 0 . 0214   0 . 5760    2  0 . 74 0 . 82   4 ( 0 . 7463 ) 0 . 0044 g g    0 . 0038 ( )      0 . 82 0 . 97 0 . 0101 g 2  g   4 ( 0 . 7789 ) 0 . 0040 g   0 . 97 g    31 . 8295 32 . 8295 g Accounts for cluster formation by multiplying the “Wen & Yu” drag correlation with a heterogeneity factor Li et al., Chem. Eng. Sci, 2012

To account for particle density distribution changes due to the reaction        f ( ξ ; x , t ) f ( ξ ; x , t ) j     [ u ( t , x ) f ( ξ ; x , t )] [ D ( ξ ; x , t ) ] [ f ( ξ ; x , t )] h ( ξ ; x , t ) p pt        t x x x t i i i j Accumulation term + Convection term + diffusive term + Growth term = Source term

Finite size domain Complete set of trial functions Method Of Moments: FCMOM      { [ ( t ) ( t )] / 2 }   min max  Finite size domain: [-1, 1] instead of [0, ∞ ]    [ ( t ) ( t )] / 2 min max  Solution in terms of both Moments and size distribution  f( ξ ,x,t) will be approximated by expansion based on a complete set of trial functions       f ( , x , t ) C ( t , x ). ( ) when n n  n 0  2 n 1 1 n ( 2 v )! 1     n v  c . . ( 1 ) . .{ }.  n 2 v n   n 2 [( 2 v n )! ] [( n v )! ].[( v )! ] 2  v 0 1 n  2 1           i f .( ) . d ( ) . P ( ) n n i 2  1          i .( . v ) ( MB MB IG ) i p Conv  t Strumendo and Arastoopour, 2008

 Implementation in Ansys /Fluent code via User Defined Scalars and Functions CFD Multiphase Model Phase velocity, Mean particle size Volume Fraction PBE terms Moments of size Reaction Distribution Population Balance Model     i           s s s i i i i .( v D ) S  s s p s s s s s  t