Modeling Aerosol Growth in Amine Scrubbing

for Carbon Capture

Yue Zhang, University of Texas at Austin

Gary T. Rochelle, University of Texas at Austin

Executive Summary• Research Objective: understand growth mechanisms and develop strategies to

remove aerosols by quantitative and accurate modeling

• Growth mechanisms:

o as part. conc increases, aerosol growth decreases due to amine driving forcedepletion

o the limiting driving force of aerosol growth is amine

o high amine volatility increases growth

• Strategies to remove aerosols:

o reduce aerosol nuclei below 106 part./cm3

o choose solvents with moderate volatility, avoid solvents with low volatility

o expand WW and pre-humidify dry bed

2

Amine Scrubbing Carbon Capture

Flue gas4~12% CO2

7% H2O

CrossExchanger

Rich Solvent

Absorber1atm40-60 oC

Stripper5 bar150 °C

Lean Solvent

Condenser

CO2 to Compressor

Reboiler

Water Wash40-50 oC

Stack

3

Amine Scrubbing Carbon Capture

Flue gas4~12% CO2

7% H2O

CrossExchanger

Rich Solvent

Absorber1atm40-60 oC

Stripper5 bar150 °C

Lean Solvent

Condenser

CO2 to Compressor

Reboiler

Water Wash40-50 oC

Stack

Nuclei for aerosolsSOx, fly ash

4

Amine Scrubbing Carbon Capture

Flue gas4~12% CO2

7% H2O

CrossExchanger

Rich Solvent

Absorber1atm40-60 oC

Stripper5 bar150 °C

Lean Solvent

Condenser

CO2 to Compressor

Reboiler

Water Wash40-50 oC

Stack

Nuclei for aerosolsSOx, fly ash

5

Bad

• if aerosol is not captured: emissions resultin solvent loss & environmental impact

Good

• if aerosol grows enough: will be captured.> 3 µm, captured by impaction

This work

• How much aerosol grows

• How we can manage aerosol growth

PZ Makes Aerosol• April 2017 UT-SRP pilot plant with 5 m PZ and 52 ppm SO3

Before SO3 injection

6

After SO3 injection

4/26/20174/26/2017

• Limiting driving force for growth

• Solvent selection

• Operating conditions

• Effective process configurations

Growth Mechanisms Are NOT Well-understood

7

Industrial Configurations for Emissions Control

Configurations demonstrated in field

• Acid Wash by Aker Solutions1,2

• Two-stage Water Wash by Linde-BASF3

• Dry Bed by BASF-Linde-RWE Power4,5

In this study

• Aerosol with a wide range of particle number conc

• Amine with different volatility

• Dry Bed, Intercoolers, Multi-stage Water WashJ. Knudsen, et al., 20131

O. Bade, et al., 20142

T. Stoffregen, et al., 20143

P. Moser, et al., 20134

P. Moser, et al., 20145

8

PZ Model by Fulk1, Kang2, and Zhang

• Steady-state absorber and water wash simulations in Aspen Plus, andaerosol calculations in gPROMS

• Proposed gas phase amine driving force depletion

MEA Model by Majeed3

• Steady-state absorber simulations in NTNU in-house simulator, andaerosol calculations in MATLAB

• Also proved gas phase MEA depletion

9

Sequential Aerosol Growth Model

Fulk, et al., 20161

Kang, et al., 20172

Majeed, et al, 20173

Sequential Aerosol Growth Model

Bulk liquid

Aerosol

CO2 (g)CO2 (aq)

Bulk gas

T, Pxi,yi

𝒑𝒓𝒐𝒑𝒆𝒓𝒕𝒊𝒆𝒔

Aerosol-gas

H2O (g) H2O (l)H2O (l)

CO2 (aq)

PZ (g)PZ (aq)PZ (aq)

User defined closed-formequations for P*

CO2& P*

PZ

10

Sequential Aerosol Growth Model

Bulk liquid

Aerosol

CO2 (g)CO2 (aq)

Bulk gas

T, Pxi,yi

𝒑𝒓𝒐𝒑𝒆𝒓𝒕𝒊𝒆𝒔

Aerosol-gas

H2O (g) H2O (l)H2O (l)

CO2 (aq)

PZ (g)PZ (aq)PZ (aq)

User defined closed-formequations for P*

CO2& P*

PZ

11

Bulk liquid

Aerosol

CO2 (g)CO2 (aq)

Bulk gas

T, Pxi,yi

𝒑𝒓𝒐𝒑𝒆𝒓𝒕𝒊𝒆𝒔

Aerosol-gas

H2O (g) H2O (l)H2O (l)

CO2 (aq)

PZ (aq)PZ (aq)

12

PZ (g)

Model Assumptions

PZ mass transfer decreases gas phase PZ

CO2, H2O, N2, T, P still remain constant

Preliminary Modeling Results

Aerosol Growth at Realistic Plant Conditions

13

14

10’

30’

20’

20’

40 oC

40 oC

National Carbon Capture Center(NCCC) Absorber

• Nov 2017 NCCC Campaign

• 0.5 MWe Pilot Solvent Test Unit

• 90% removal

Rate-based Absorber Modeling

• Independence Model

• Developed in Aspen Plus® RateSepTM,1

• Rigorous e-NRTL thermodynamic framework

• Rigorous kinetics with reactions in boundary layer

• Solvent

• 5 m PZ: fast absorption rate, low viscosity, good energy performance

• Lean Loading at 0.22 (mol CO2/mol alk)

P. Frailie, 20141

15

Assumptions for Aerosols

• Well-mixed

• Particle conc at 107 part./cm3

< 106 : emits < 1ppm amine

> 108 : starts coagulation

~ 107 : most often observed at site

• Initial conditions

0.1 µm, 5 m PZ, 0.36 CO2 loading

H2O (l)

CO2 (aq)

PZ (aq)

16

0

2

4

6

8

10

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Dia

met

er (

µm

)

Z/Ztot

WWABS Dry Bed

General aerosol growth profile

• Aerosols grow from 0.1µm to 4.4 in ABS, and 10 in WW (collectable)

• Aerosol initial diameter is not critical

107 part./cm3

17

1.E-07

1.E-06

1.E-05

1.E-04

1.E-03

1.E-02

1.E-01

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Na

no

mole

Z/Ztot

ABS WW

CO2

PZ

H2O

Dry Bed

Component pickup in aerosol

• Aerosols grow in WW by picking up water

107 part./cm3

18

0

1

2

3

4

5

6

7

8

9

10

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Dia

met

er (

µm

)

Z/Ztot

1 part./cm3

107

5*107

108

ABS Dry Bed WW

High part. conc reduces aerosol growth

19

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Gas

Ph

ase

PP

Z(P

a)

Z/ZtotABS BOT ABS TOP

1 part./cm3

107

108

PZ driving force depletion in gas phase

20

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Gas

Ph

ase

PP

Z(P

a)

Z/ZtotABS BOT ABS TOP

1 part./cm3

107

108

P*pz, drop

P*pz, l

PZ driving force depletion in gas phase

21

Relative driving force ratio between g-d

• The limiting driving force of growth is PZ

• As part. conc increases, limiting driving force (PZ) shifts from g-d to l-g

• Aerosol is always in equilibrium with water in gas

P*lPg

P*d Pg

∅𝑔𝑑 = 100%

∅𝑔𝑑 = 0%

1 part./cm3 107 108

Avg ∅𝑔𝑑, PZ 100% 72% 32%

Avg ∅𝑔𝑑, water 0% 0% 0%

bulk liquid

22

0

2

4

6

8

10

12

14

16

18

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Dia

met

er (

µm

)

Z/Ztot

ABS WW 40 oC WW

40 oC

Dry

Bed

(Dry Bed)

Base Case

Dry bed needs to be pre-humidified to grow aerosols

107 part./cm3

23

ABS

0

2

4

6

8

10

12

14

16

18

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Dia

met

er (

µm

)

Z/Ztot

WWDry Bed

10x PPZ

0.1x PPZ

Base Case

• Choose solvents with moderate volatility, like PZ (collectable)

• Avoid solvents with low volatility (non-collectable) 24

Increase and decrease PZ volatility by 10x

107 part./cm3

Conclusions - growth mechanisms

• As part. conc increases, aerosol growth decreases due to amine driving

force depletion. The limiting driving force shifts from g-d to l-g

• In NCCC with 5 m PZ

o 107 part./cm3 are collectable

o w/o water wash, 108 part./cm3 are non-collectable

• In water wash, aerosol grows by picking up water

• Higher amine volatility increases growth

25

Recommendations

• Nuclei

o Reduce aerosol nuclei below 106 part./cm3

• Solvent selection

o Choose solvents with moderate volatility, like PZ

o Avoid solvents with low volatility

• Process configurations

o Expand WW

o Pre-humidify dry bed

26

Possible Future Work

• Test a wide variety of amines/blends

• Particle size/residence time distribution

27

For more information

Yue Zhang, Ph.D. Candidate, University of Texas at Austin

yuezhang1992@utexas.edu

Gary T. Rochelle, Professor, University of Texas at Austin

gtr@che.utexas.edu