NETL 2012 Conference on Multiphase Flow Science

Javad Abbasian, Emadoddin Abbasi and Hamid Arastoopour

Sulfur By-Product

Fly Ash By-Product

Slag By-Product

0.01

0.1

1

10

100

250 300 350 400 450 500Temp, C

PC

O2,

atm

AbsorptionMgO+CO2 MgCO3

DecompositionMgCO3 MgO+CO2

°

Advanced Power Plant

CO2 Capture Reactions MgO + CO2 MgCO3 Absorption MgO + CO2 MgCO3 Regeneration

Water-Gas-Shift Reaction CO + H2O CO2 + H2

0

5

10

15

20

25

30

35

40

0 2 4 6 8 10 12 14 16 18Time, min

Con

vers

ion,

%

350 C

500 C

425 C

450 CP= 20 barCO2/N2: 50/50Sorbent: EP44

4

CO2 Capture

0

20

40

60

80

100

0 5 10 15 20 25Time, min

Dec

ompo

sitio

n, %

550 C

500 C

450 CP= 20 barN2: 100%Sorbent: EP44

Sorbent Regeneration

20 µm Scale:

0

5

10

15

20

25

30

35

40

0 5 10 15 20 25 30Time, min

Conv

ersi

on, %

With Potassium Impregnation

Without Potassium Impregnation

200 µmScale: 200 µmScale: 200 µmScale:

Mass balance around the particle for reactant gas (i.e. CO2)

B.C.: (a) @ R = 0 (b) @ R = Rp Mass balance around the grain for reactant gas (i.e. CO2) B.C.: (a) @ r=rg’ (b) @ r=ri

( ) ( ) 01313

22

2 =−−−

∂∂

∂∂ n

eig

iRe CCk

rr

RCRD

RRε

01 22 =

∂

∂

∂∂

rC

rrr

D gg

Rg CC = ( )negg

g CCkr

CD −=

∂

∂.

Rg CC =

Radius at Reaction interface

I.C.: ri = rg @ t=0 Overall Conversion of the particle

( )egMgO

i CCkdtdr

−−=ρ

dRrrR

RX

R

g

ioverall ∫

−=

0

03

32

30

.1.3

0=∂∂

RCR

7

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0 5 10 15 20Time, min

MgO

Con

vers

ion

TGA Experimental DataTwo-Zone Reactive ModelOne-Zone Reactive ModelUnifrom Reactivity Model

R2=0.99R2=0.97R2=0.90

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0 5 10 15 20Time, min

Con

vers

ion

500 °C

450 °C

425 °C

350 °C

10

0

10

20

30

40

50

60

70

80

90

100

0 10 20 30 40 50 60 70 80 90Time, min

Mol

ar P

erce

nt (M

ole

of C

O2

out/M

ole

of C

O2

in)

350°C

300°C

400°C

425°C

0

10

20

30

40

50

60

70

80

0 10 20 30 40 50 60 70 80 90Time, min

Mol

ar P

erce

nt (d

ry b

ase)

CO2

H2

CO

- For heterogeneous reactive systems: Conservation of species in each phase

- Rj is the heterogeneous reaction rate

- An Ideal reaction model for CFD applications should: - Capture the real physics - Not increase the CPU time significantly - Be easy to implement in CFD codes

- Grain Model:

- Captures the real physics - Is an implicit model (needs 2 additional equations ) - Increases CPU time - Is not easy to implement in commercial CFD codes

jigggigg Ryvyt

=∇+∂∂ ).()( ρερε

Numerical Modeling Species Conservation Equations

Mass balance around the particle for reactant gas (i.e. CO2)

B.C.: (a) @ R = 0 (b) @ R = Rp Mass balance around the grain for reactant gas (i.e. CO2) B.C.: (a) @ r=rg’ (b) @ r=ri

( ) ( ) 01313

22

2 =−−−

∂∂

∂∂ n

eig

iRe CCk

rr

RCRD

RRε

01 22 =

∂

∂

∂∂

rC

rrr

D gg

Rg CC = ( )negg

g CCkr

CD −=

∂

∂.

Rg CC =

Radius at Reaction interface

I.C.: ri = rg @ t=0 Overall Conversion of the particle

( )egMgO

i CCkdtdr

−−=ρ

dRrrR

RX

R

g

ioverall ∫

−=

0

03

32

30

.1.3

0=∂∂

RCR

rc : Radius of the low reactive zone (k2) rp : Initial radius of the particles rp’: Radius of the expanded particle Low Reactive Zone (k2)

Highly Reactive Zone (k1)

Product Layer

Gas Film

rp, k1

rc k2

rp’

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0 200 400 600 800 1000 1200 1400 1600

MgO

Con

vers

ion

Time (s)

Exp 425°C

Variable Dg

SCM

* Abbasi et al., Fuel , 2012

1- There are two distinct reactive zones inside the particles 2- Process is controlled by both surface reaction and product

layer diffusion 3- There is an Expanding product layer 4- Effective product layer diffusivity is a function of conversion

due to expansion ( and increased tortuosity ) 5- Intrinsic reaction rate is Arrhenius type

)exp(0 RTEkk ss −=

)(0βαXDD gg −=

3 )1( ZXXrr pp +−′=

action

s

Layeroduct

gp

iip

gp

i

oMgOebi

kDrrrr

krr

NCCdtdr

RePrfilm

2 1)(/)(

+−

+

−=−

3)(1p

i

rr

X −= dXXr

dr pi

32

)1(3

−−−=

3 )1( ZXXrr pp +−′=

productreact

reactproduct

MM

Z⋅

⋅=ρρ

′−+

−−=

)1(1

)(

p

ii

g

s

eboMgO

si

rr

rDk

CCN

kdtdr

)

111()1(1

)1)((3

331

32

XZXXXr

Dk

XCCN

kr

dtdX

pg

s

eboMgO

s

p

+−−

−−+

−−

−=rp, k1

rc,k2

c

cs rrfork

rrforkk

<≥

=

2

1

)(0βαXDD gg −=

------------------------------------------------------------------------------------------------------------------------------------------- Onischak and Gidaspow, ”Separation of Gaseous Mixtures by Regenerative sorption on Porous Solids. Part II: Regenerative separation of CO2”, Recent Developments in Separation Science, ed. N. Li, 1972

0

5

10

15

20

25

30

0 500 1000 1500 2000 2500 3000 ϕ(

t)

t(s)

Thiele Modulus in our study

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0 500 1000 1500

MgO

Con

vers

ion

Time (s)

Exp 350 °C

Exp 425°C

Exp 450°C

Model

0

1E-10

2E-10

3E-10

4E-10

5E-10

6E-10

0 0.5 1

Dg

(m2/

s)

MgO Conversion

Model

Parameter

Without Steam With Steam

350 o C 425

o C 450

o C 425

o C

Dg0 1.1E-10 6.0E-10 1.2E-9 2.1E-9

α 3.5E+2 8.1E+1 5.7E+1 4.6E+1

β 3 3 3 3

)( 30 XDD gg α−=

45 cm

30 cm

Thermocouple

Thermocouple

8 cm

7 cm

T1

Thermowell

Bed of Sorbent

Quartz Beads

SS Annulus

SS Annulus

Frit

Reactor Body

T2

T1

T2

420

425

430

435

440

445

450

455

0 5 10 15 20 25 30 35 40 45 50Time, min

Tem

pera

ture

, C

T11T12T13T14

420 425 430 435Bed Temp., C

Posi

tion

of th

e be

d, c

m

0

8

4 8 cm

2 cm

T11

T12

T14

T13

Thermowell

420 425 430 435Bed Temp., C

Posi

tion

of th

e be

d, c

m

0

8

4 8 cm

2 cm

T11

T12

T14

T13

Thermowell

@ 425°C start 50 min

CO2 Absorption Breakthrough Curve at Different Operating Temperatures

0

0.5

1

1.5

2

2.5

3

3.5

4

0 5 10 15 20 25 30 35 40Time, min

CO

2 Out

let F

low

, mol

/min

*103

Sorbent: EP68System Pressure= 20 atmInlet Total Flow Rate= 200 cm 3 /minCO2 inlet= 50 %molN2 inlet= 50 %mol

450°C

425°C

350°C

Bed Inlet

Bed Outlet

An explicit reaction rate model has been developed which is able to describe the reaction rate data obtained in TGA.

The reaction model is suitable for CFD applications due to its explicit formulation.

The reaction model was validated vs the experimental data obtained in pack-bed.

Future work includes:

Application of the CFD and reaction models in simulations of a bubbling bed and circulating fluidized bed reactors.

Modeling of Regeneration reaction

Modeling of the pr0cess

Financial support for this study was provided by the U.S. Department of Energy (contracts DE-FG26-03NT41798 and DE-FE0003997) and the Illinois Clean Coal Institute (ICCI Project No: 11/US-4).