Post-combustion CO2 capture by Ca-loopingmathematica.nodo.cesga.es › combustion › docs › Slides_Arias.pdf · MODELLING. Particle models. OD/1D Models. 2D/3D Models. Post-combustion

"Post-combustion CO2 capture by Ca-looping"

Borja Arias RozadaCO2 Capture Group

National Institute of Coal (INCAR-CSIC)

Workshop on Mathematical Modelling of Combustion

23-25 May, Santiago de Compostela, Spain

Post-combustion CO2 capture using Ca-looping

• What is Ca-looping?• Ca-looping for post-combustion CO2 capture: description of the process• Status of the technology

• Modelling work:• Particle models• Reactor models

Post-combustion CO2 capture using Ca-loopingWorkshop on Mathematical Modelling of Combustion

0,001

0,010

0,10

1,00

10,00

100,0

600 700 800 900 1000 1100

Carbonation

CalcinationCaO + CO2 CaCO3

Lime

CO2 capture

Gas with CO2

Sorbent regeneration

Sorbent+CO2CO2

Fresh sorbent

Sorbent

Spent sorbent

Energy

CO2 capture using solid sorbents

What is a Ca-looping?

Use of lime as sorbent

LimestoneP C

O2

eq, a

tm

Temperature (ºC)


Advantages of Ca-looping versus other post-combustion technologies

CARBONATORCOMBUSTOR

Concentrated CO2

CoalCaCO3(F0)

CaO Purge

CaCO3

CaOFlue Gas

Flue gas“without” CO2

Coal Air

O2

Air

N2

ASU

CALCINER

Power OUT

Power OUT

CO2

Some advantages of Ca-looping:• Low energy penalty

• Purge of CaO: synergies with cement industry

• Pre-treatment of flue gas no needed Post-combustion CO2 capture using Ca-loopingWorkshop on Mathematical Modelling of Combustion


Technical development of Ca-looping

• Classification of CO2 capture technologies:• Near commercial (amines, oxy-fuel combustion, pre-combustion)• Emerging technologies

Roadmap proposed for the development post-combustion CO2 capture with Ca-looping

Post-combustion CO2 capture using Ca-looping

• What is Ca-looping?• Ca-looping for post-combustion CO2 capture: description of the process• Status of the technology

• Modelling needs:• Particle models• Reactor models



EXPERIMENTAL

MODELLING

Particle models OD/1D Models 2D/3D Models

Post-combustion CO2 capture using carbonate looping

Lab ScaleSmall pilot plant scale

Pilot plant scale (1.7 MWt) Demo ScaleScale

CO2

CaCO3

CaSO4

CaO

Conversion (X)

Reactivity (dX/dt)

SO2


0.00

0.35

0 50 100 150 200

t (s)

X CaO

1

2

Particle Models: CaO conversion to CaCO3Ca

O c

onve

rsio

n

Time (s)

Typical conversion of CaO particles to CaCO3

( )(1 ) 1 ln(1 )

(1 ) 1 1 ln(1 ) 1

− − Ψ −=

− + − Ψ − − Ψ

s o s

o

k S C X XdXZdt Xβ

ε

Maximum CO2 carrying capacity under the chemical fast reaction regime

Cycle 1Cycle 2

Cycle 3Cycle 4Cycle 10

Cycle 20

CaO

con

vers

ion

Time (s)

Cycle 1Cycle 2

Cycle 3Cycle 4Cycle 10

Cycle 20

CaO

con

vers

ion

Time (s)

1. Chemically controlled

CaO

CO2

2. Diffusion/Chemically controlled

CaO

CO2

CaCO3

Reaction rate kinetics: Random Pore Model

Ref: Grasa et al. AIChE Journal, 2009, (55) 1246-1255

Effect of number of cycles on sorbent conversion

XN

XN

Reactivity fast enough for circulating fluidized beds

kso (m4/kmol.s) Eak (kJ/kmol) Do (m2/s) EaD(kJ/kmol)

0.54*10-5 20.3*103 3.85*10-6 160*103

Kinetic parameters

Changes of textural properties during cycling

Sintering of the sorbent


Average activity of the solid in a Ca-looping

Ref: Grasa et al. Ind. Eng. Chem. Res. 2006, 45:8846-8851Abanades, Chemical Eng. Journal 2002, 90: 303-306Rodríguez et al. Chemical Eng. Journal, 2010, 156: 388-394

NR

NR

N FFFFr

)( 0

10

++

=−

Sorbent decay conversion

Continuous system: Particles with different number of cycles

∑∞

=

=1N

NNave XrX

Average activity of the particles in the system

Rapid decay during initial cycles Residual conversion

(0.07-0.10) after hundreds of cycles

r

r

N XNk

X

X ++

−

=

11

1

Other aspect to consider:•Presence of SO2 in the Ca-looping•Partial carbonation and calcination of the sorbent

F0

F0

FR

FR

Fgas+FCO2

Fgas FCO2

CalcinerCarbonator

Carbonation: 650 ºCCalcination:850 ºC

K=0.52Xr=0.075

0

0.1

0.2

0.3

0.4

0.5

0.00 0.05 0.10 0.15 0.20

FO/FRX a

ve

Effect of Fo on Xave

Strategies of operation:•Low Xave and high FR•High Xave and low FR

FCO2 capt = FR * Xave

EXPERIMENTAL

MODELLING





CO2

CaCO3

CaSO4

CaO

Conversion (X)

Reactivity (dX/dt)

SO2


Small pilot-plant facility: Description

Small pilot plant at INCAR-CSIC (30 kWt)

Two interconected circulating fluidized beds

Carbonator reactor

CARB

ON

ATO

R

CALC

INER

Simulatedflue gas (CO2)

AirCoal

Clean flue gasCO2


Main features:

• Two CFB reactors (Height~6.5 m, diameter=100 mm)

• Electrically heated

• Measurement port (temperature, pressure, gas composition)

• Solid circulation measurements

• Solid samples characterization (TG analysis, C/S analyzer)

Ref: Rodriguez et al. AIChE J 2011, 57 (5) 1356-1366

Steady state:Defined as the situation where carbonator and calciner temperature, pressuredrops, inlet gas flows and outlet gas concentration remain constant for aperiod of time of at least 10-20 minutes.

Some measured experimental parameters:•Average carbonation temperature•Inventory of solids•Inlet CO2 concentration and total flow•Outlet CO2 concentration•Carbonate content of entering and exiting solids•Average CO2 carrying capacity of solids•Solid circulation rates

Small pilot-plant facility: Experimental results

Main achievements:

• Has completed 450 h of operation

• Max. CO2 capture efficiency: 97 %

• Absorption capacity up to 7 molCO2/m2s)

0

20

40

60

14:05 14:15 14:25 14:35 14:45 14:55 15:05Inve

ntor

y of

sol

ids

(cm

H2O

)

Calciner Carbonatorr

0

5

10

15

20

25

14:05 14:15 14:25 14:35 14:45 14:55 15:05

Con

cent

ratio

n (%

)

CO2O2O2 probe

QT= 19 m3/h, νCO2 = 0.12, Xave= 0.08

500

600

700

800

900

1000

14:05 14:15 14:25 14:35 14:45 14:55 15:05

Tem

pera

ture

(ºC

) Carbonator Calciner

CO2~2.5%


Description of the IFK (University of Sttutgart) 10 kWth facility

Ref: Charitos et al. Int. J. Green. Gas. Con., 2010, (4) 776-784

Main features:• A CFB (height=12m) and a BFB• Electrically heated• Measurement ports (temperature, pressure, gas composition)• Solid circulation measurements• Cone vale to control sorbent flow between reactors• Posibility of oxy-fuel combustion in the BFB calciner

Model validation in two experimental facilities: IFK and CSIC



FCa Xcalc from calciner

FCa Xcarb to calcinerFCO2 from flue gas

FCO2 (1-Ecarb)

Carbonator reactor concept

Initial assumptions carbonator reactor:•Instantaneous and perfect mixing of the solids•Plug flow for the gas phase

Overall mass balance in the carbonator

Carbonator reactor

NCaO

Xave

in CO2

in CO2inCO2carb F

FF)(E efficiencyn Carbonatio −=

CO2 reacting with CaO in the bed

CO2 removed from the gas phase

CaCO3 formed in the circulating stream of CaO= =

Ref: Rodriguez et al. AIChE J 2011, 57 (5) 1356-1366

0

20

40

60

80

100

14:05 14:25 14:45 15:05

Eca

rb (%

)

0

0.03

0.06

0.09

0.12

0.15

νCO

2

Ecarb Ecarb_eq CO2


in CO2

in CO2inCO2carb F

FF)(E efficiencyn Carbonatio −=

Continuous measurement of Ecarb during an experimental run:

•Flue gas fed to the carbonator

•CO2 concentration at the inlet and exit of the reactor

CO2 mass balance in the systemCO2 reacting with

CaO in the bedCO2 removed from

the gas phaseCaCO3 formed in the

circulating stream of CaO= =

Tcarb = 655 ºC, WCa = 260 kg/m2, nCO2 = 0.12, Xave= 0.08

CARB

ON

ATO

R

CALC

INER


AirCoal

Clean flue gasCO2

Ftotal

CCO2 in

CCO2 out




Experimental measurement

CARB

ON

ATO

R

CALC

INER


AirCoal

Clean flue gasCO2

Supply of active flow of CaO

XcalcXcarb

FCa

FCO2

( )calcaveCacarbCO XXFEF2

−≤





Ref: Rodriguez et al. Energy Procedia 2011, (4) 393-401

( ) 2COcarbcalccarbCa3 FEXXF

CaOofstreamgcirculatintheinformedCaCO

=−=



( )e2COavesave ffXkr −=





Particles reacting in the carbonator

>

<=

*tfor t 0

*t tif *t

Xr

N

N

>

<=

*tfor t 0

*t tif *t

Xr

ave

ave

Simplified reaction rate

aCaActiveCa fNN =

−= τ

−*t

a e1f

CaOFCaN

=τt*

Xave

Ref: Alonso et al. Chem. Eng. Journal, 2009, (64) 883-891

( )e2COavesaCareactor

ActiveCa ffXkfNdtdXN

bedtheinCaOwithreacting2CO

−ϕ=

=


Apparent reaction rate=0.43 s-1

0.0

0.2

0.4

0.6

0.8

1.0

0 50 100 150 200

faNCaXave, mol/m2

Ecar

b

ExperimentalModel

Comparison of CO2 capture efficiencies and CO2 removal rates in the reactor






( )2CO

e2COavesaCacarb F

ffXkfNE

−ϕ=

FCO2= 7 mol/m2s

The CO2 capture in the reactor is controlled mainly by the kinetics of the carbonation reaction

Experimental determination of ks

ϕ close to unity

aveaCO

CaOactive Xf

FN

2

=τ

Experimental data:

Inlet CO2 Vol. % = 11.4 for IFK &16.5 for INCAR-CSIC

Tcarb= 634-660 °C, Xmax,ave= 0.08-0.23

Ref: Charitos et al. Ind. Eng. Chem. Res. Submitted for publication. Common paper between IFK and INCAR-CSIC

Model validation in two experimental facilities: IFK and CSIC

Critical active space time at which

ECO2/Eeq>90%:

• IFK data set: 0.025 h

• INCAR-CSIC data: 0.008 h

Main differences between both series of experimental and calculated data• Different inlet CO2 concentration (11.4% IFK, 16.5% INCAR)

• Different limestone (IFK limestone less reactive, ks=0.20s-1)

Active space time

( )e2COsactivecarb ffkE −ϕτ=( )

2CO

e2COavesaCacarb F

ffXkfNE

−ϕ=

EXPERIMENTAL

MODELLING





CO2

CaCO3

CaSO4

CaO

Conversion (X)

Reactivity (dX/dt)

SO2


Next steps in Ca-looping development: 1.7 MWt pilot plant

Facility area

FP7 “CaOling” Project:Development of postcombustion CO2 capture with CaO in a large testing facility: “CaOling”

Main objective:

To advance in the experimental validation of the carbonate looping cycle (at a 1 MWt scale range) and demonstrate that this is a low cost, highly energy efficient CO2 capture technology, suitable for retrofitting coal combustion power plants

Operation will start in September 2011



Calciner operating under air or oxy-combustion

conditons

Double loop seal to control solid circulation

between reactors

Removable cooling bayonet tubes to change

heat extraction in reactors

O2 and CO2 tanks and gas mixer, to change oxy-combustion ratio

Calciner gas heater and indepent steam inyection

Gas velocities similar to those encountered in CFB

boilers (3-5 m/s)Continuous coal and

limestone feeding system

Next steps: 1.7 MWt pilot plant

Complete instrumentation (temperature, pressure,

solid samples)

Next steps: 1.7 MWt pilot plant


SOME OBJECTIVES:• To gain the necessary design data and experience for rapid scale-up of the technology, building

experimentally validated models from the careful interpretation of results produced from lab-scaleprototypes and from the experimental campaigns in a 1 MW test facility.

• To evaluate and optimise the concept in operating conditions equivalent to large-scale industrial unitsand integrated in a commercial plant. To analyse the controllability and stability of the process

• To find the optimum set of operating conditions to minimize sorbent make-up flow cost (calcinationtemperatures, O2/CO2 ratios in the calciner, requirement for steam in the carbonator and calciner, bestsuitable approach for SO2 capture, etc).

• Modelling work. Next steps:• Solid distribution of the whole system: pressure balance approach

(inventory of solids in the reactors, solid circulation rates,…)• 2D/3D reactor models including the more complex fluid-dynamics of large

scale systems• CFB Calciner reactor: oxy-fuel combustion+calcination

Postcombustion Calcium Looping is a rapidly developing technology successfullycharacterized at small pilot plant scale.

A simple carbonator reactor model (CSTR for solids and PF for gas) is shown to fitreasonably well the available results in CFB mode of operation.

The model highlights the importance of sufficient solid inventories and solidcirculation rates for a given activity of the solids and a given flow of flue gasesentering the carbonator.

The extrapolation of results to large scale CFB carbonator reactors anticipates captureefficiencies over 90% in operating at realistic conditions (similar to those present incommercial CFBC units) even with CaO particles in their residual activity after 10s to100s of carbonation-calcination cycles.

Carbonation efficiency correlates well with active space time, which combines themain carbonation operation parameters.

A flexible experimental facility has been constructed in La Pereda Power Plant (Spain)aiming to validate Calcium Looping technology in the 1MWs size.

Conclusions/Remarks


"Post-combustion CO2 capture by Ca-looping"

Workshop on Mathematical Modelling of Combustion

23-25 May, Santiago de Compostela, Spain

Thank you for your [email protected]

•This work has been carried out as part of the FP7 “CaOling“ Project.•Special thanks to IFK (Prof. Scheffknecht) for permission to use some of their data and figures in this presentation.

Post-combustion CO2 capture by Ca-loopingmathematica.nodo.cesga.es › combustion › docs › Slides_Arias.pdf · MODELLING. Particle models. OD/1D Models. 2D/3D Models. Post-combustion

Documents