www.ecn.nl

Advanced Membrane Reactors for Carbon-Free Fossil Fuel Conversion

Daniel Jansen, Wim Haije, Michiel Carbo, Virginie Feuillade, Jan Wilco Dijkstra,Ruud van den Brink

Contents

The project

Objectives

R&D results ECN

Reactor design

System analysis

Materials science

Catalyst testing

Conclusions

Advanced Membrane Reactors in Energy Systems Development of novel membranes for membrane reactors.

Objective:

The purpose of this project is to develop H2 and CO2 membranes to allow combinations of natural gas reforming or WGS with H2 or CO2 separation in separation enhanced reactors, i.e. membrane reactors, for carbon-free hydrogen production or electricity generation.

systemstudies

reactordesign

membrane & catalyst

development

materialsresearch

experimentalresults

reactorrequirements

desired specificationsfundamental knowledgecharacterization

overall efficiencieseconomics

reactor testspatentsIP

IPpublications

newdevelopments

Task 1. System analysis and thermodynamic evaluations Task 2. Hydrogen membrane research & development Task 3. CO2 membranes research & development Task 4. Catalyst screening Task 5. Reactor modelling and design

Executed by ECN Executed by TUD Executed by ECN+TUD Executed by ECN Executed by ECN

The ECN GCEP project layout

Application:

NGCC with CO2 membrane reformer reactor

Advantages compared to H2 membrane reformer:• eliminating the requirement of water gas shift reactors: cost reductions • offering higher conversion efficiencies at lower temperatures • H2 rich stream remains at elevated pressure and temperature • CO inhibition of membrane not foreseen• no need for CO2 cleaning section

Steam reformermembrane reactor

Natural gas

H2O

Pre-reformer

Steamsweep

H2/

CO2

Steam Reforming

Combined Cycle

Application:

IGCC with CO2 membrane water gas shift reactor

Advantages compared to H2 membrane WGS reactor:• eliminating the requirement of LT water gas shift reactor: cost reduction• incomplete CO conversion in WGS does not reduce the IGCC efficiency but lowers CO2 capture ratio• H2 rich stream remains at elevated pressure and temperature• CO inhibition of membrane not foreseen• no need for CO2 cleaning section

Water gas shiftmembrane reactorCoal

H2O

G-E gasifier

Steamsweep

O2

Pre-shift

G-E product gas(around 1300 oC)

CoolerT=300 oC

H2O

Combined Cycle

CO2

Main Question

Membrane application:‘Everybody’ agrees on the fact that either H2 or CO2 selective membranes are viable options for carbon capture technologies

But……..:‘Nobody’ cares or dares to look into the process boundary conditions:Are both membranes equally fit to operate in a certain process??

Reactor Design/System analysisCO2 versus H2 selective membranes

Membrane reformer: Residual partial pressure of permeating compontent in retentate as a function of conversion.

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

0 10 20 30 40 50 60 70 80 90 100

Overall conversion (%)

H2 a

nd C

O2 r

eten

tate

par

tial

pres

sure

(bar

)

CO2 membrane

H2 membrane

Steam reformermembrane reactor

Natural gas

H2O

Pre-reformer

Steamsweep

H2/

CO2

Steam Reforming

Combined Cycle

Conversion relatively easy enhanced by separation of H2 favourable kinetics

CO2 selective membranes show a too low conversion and are therefore not suitable, as opposed to H2 selective membranes

Reactor Design/System analysis

CO2 versus H2 selective membranes: Reforming

@ 600 oC, 40 bar, S/C = 3,Sweep: steam 5 bar, 600 oC, Sweep flow/Feed flow = 0.11 (mole/mole) See poster

0,0

5,0

10,0

15,0

20,0

25,0

70 75 80 85 90 95 100

WGS conversion (relative to product gas) (%)

Part

ial p

ress

ure

H2 a

nd C

O2

in re

tent

ate

(bar

)

H2 membrane

CO2 membrane

Water gas shiftmembrane reactorCoal

H2O

G-E gasifier

Steamsweep

O2

Pre-shift

G-E product gas(around 1300 oC)

CoolerT=300 oC

H2O

Combined Cycle

Water gas shiftmembrane reactorCoal

H2O

G-E gasifier

Steamsweep

O2

Pre-shift

G-E product gas(around 1300 oC)

CoolerT=300 oC

H2O

Combined CycleCombined Cycle

H2 permeation results in better CO2recovery and CO conversion but CO2permeation does not perform much worse

Complete system and exergy analysis needed to really determine all pros and cons

Reactor Design/System analysis

CO2 versus H2 selective membranes: IGCC

CO2

@300 oC, 42.2 bar, S/C just enough to complete water gas shift, excess catalyst, Sweep: steam 17 bar, 300oC See poster

Reactor Design/System analysisIGCC with H2 WGS Membrane Reactor

N2H2, CO,

H2O, CO2

Slag

PulverizedCoal

Quench Gas

BFW

IP Steam

HP Steam

Syngas

H2

H2, N2

O2

Air

N2

N2

Air

CO2

Stack

H2-WGSMembrane

Reactor

ASU Dry-fed CoalGasifier

GasCooler

GasTurbine

HRSG SteamTurbine

CO2Liquefaction

WGSGasCleaningSection

IP SteamH2O

IP Steam

CryogenicDistillation

CO, H2, N2, CH4

Reactor Design/System analysisIGCC with CO2 WGS Membrane Reactor

CO2, H2O

H2, CO,H2O, CO2

Slag

PulverizedCoal

Quench Gas

BFW

IP Steam

HP Steam

Syngas

H2, N2

O2

Air

N2

N2

Air

CO2

Stack

CO2-WGSMembrane

Reactor

ASU Dry-fed CoalGasifier

GasCooler

GasTurbine

HRSG SteamTurbine

CO2Liquefaction

WGSGasCleaningSection

IP Steam

H2OIP Steam

CO2

LP Steam

H2

Reactor Design/System analysisCO2 versus H2 selective membranes: Exergy analysis IGCC

22500

15000

-

-

MembraneArea[m2]

55.6

-

-

-

SweepSteam[kg/s]

38.1

40.7

39.9

47.9

Efficiency

[-]

85.5

84.8

91.7

-

CarbonCapture

[-]

398IGCC CO2-selective WGS-MR

425IGCC H2-selective WGS-MR

417IGCC Selexol (HT- & LT-WGS)

500IGCC Base Case

Output

[MWe]

Case

22500

15000

-

-

MembraneArea[m2]

55.6

-

-

-

SweepSteam[kg/s]

38.1

40.7

39.9

47.9

Efficiency

[-]

85.5

84.8

91.7

-

CarbonCapture

[-]

398IGCC CO2-selective WGS-MR

425IGCC H2-selective WGS-MR

417IGCC Selexol (HT- & LT-WGS)

500IGCC Base Case

Output

[MWe]

Case

0

0,5

1

1,5

2

2,5

3

3,5

4

CO2compression

CO-Shift Lost GT-work CO2 separation

Effic

ienc

y re

duct

ion

(% p

oint

s)

Losses due to CO2 capture

Sensitivity analysis on CO2 permeation: see poster

Reactor Design/System analysis: Conclusions

Reactor modeling• Membrane reformer for CO2 capture in NGCC

- CO2 selective membranes show a too low CH4 conversion and are therefore not suitable for CO2 capture in NGCC power plants

• Water gas shift membrane reactor for CO2 capture in IGCC

- CO2 selective membranes give CO conversions in IGCC comparable to H2separating membranes and offer therefore a viable alternative

System analysis IGCC- Coal gasifiers always produce H2/CO2-ratios higher than unity, resulting in a higher H2 partial pressure, which is beneficial in membrane permeation

- The steam sweep flow applied in CO2-selective WGS-MR results in higher efficiency penalty for CO2 capture compared to H2-selective WGS-MR

Materials Science-Catalysis

Stability test of commercial catalyst

0

10

20

30

40

50

60

70

0 20 40 60 80 100Time [hr]

CH

4 C

onve

rsio

n [%

]

Ni-catalystVendor A

Noble Metal catalyst Vendor B

Noble metal catalyst Vendor C

Noble metal catalystVendor A

Noble metal catalystECN

CH4 2.9%

H2O 17.5%

N2 79.6%

Flow 25 sccm

T = 500 °C

P = 1 atm

Hydrotalcite general formula: Mg6Al2(OH)16CO3·4H2O

OH-

H2OCO3

--

Mg/Al

Rhombohedral system:

a=b≈3Å c≈23Å

mR3−

CO2 transport channel??

Materials Science-assumption

SEM-EDX result hydrothermally synthesized sample (starting materials: 90/10 Mg/Al-nitrate):

Impurity phase

Main phase

Impurity phase

Materials Science-composition

ND very suitable to see

a) Light elements C, O, H

b) Difference between Mg and AlDiffraction experiment on GEM at ISIS, UK

25/75

50/50

90/10

Materials Science-composition

Materials: • Rather poor crystallinity, esp. impurities • 50/50 sample relatively pure: refinable (GSAS)

Results:• Composition Mg0.64Al0.36(OH)2(CO3)0.18·1.0H2O• Mg< 0.64: Boehmite impurity (Al rich)• Mg> 0.64: Hydromagnesite impurity (Mg rich)

Materials Science-composition

01-089-0460 (C) - Hydrotalcite, syn - (Mg0.667Al0.333)(OH)2(CO3)0.167(H2O)0.5 - Rhombo.H.axes - a 3.04600 - b 3.04600 - c 22.77200 - alpha 90.000 - beta 90.000 - gamma 120.000 - Primitive - R-3m (166)File: MG 50 pellet RT N2 after heattreatments.raw - Start : 5.000 ° - End: 80.000 ° - Step: 0.050 ° - Step time: 4. sFile: MG 50 pellet 450 N2 H2O.raw - Start: 5.000 ° - End: 80.000 ° - Step: 0.050 ° - Step time: 2. sFile: MG 50 pellet 400 N2 H2O.raw - Start: 5.000 ° - End: 80.000 ° - Step: 0.050 ° - Step time: 2. sFile: MG 50 pellet 350 N2 H2O.raw - Start: 5.000 ° - End: 80.000 ° - Step: 0.050 ° - Step time: 2. sFile: MG 50 pellet 300 N2 H2O.raw - Start: 5.000 ° - End: 80.000 ° - Step: 0.050 ° - Step time: 38. sFile: MG 50 pellet 100 N2 CO2.raw - Start: 5.000 ° - End: 80.000 ° - Step: 0.050 ° - Step time: 2. s

Lin

(Cps

)

10

100

200

300

400

500

600

700

2-Theta - Scale8 10 20 30 40 50 60 70

20oC

450oC

400oC

350oC

300oC

100oC

There is no structural memory effect!!!

In-Situ XRD under N2/CO2/H2O

Interlayer water out: ≈1.5 Å

Materials Science-composition

Hydrotalcite TGA/MS

0

10

20

30

0 1000 2000 3000 4000 5000 6000

time (s)

wei

ght (

mg)

-0,002

0,008

0,018

sampleH2OCO2

310°C 450°C100°C 240°C

1. Adsorbed water @100oC

H2O(ad)↔H2O(g)

2. Interlayer water @240oC

H2O(abs)↔H2O(g)

3. Hydroxides @310oC

Mg(OH)2 ↔MgO + H2O(g)

4. Carbonates @450oC

Mg(OH)2 + MgCO3 ↔2MgO +CO2(g) + H2O(g)

First tentative working hypothesis

Materials Science-material stability

Dense membrane produced using cold isostatic pressing under 2000 bar sliced to about 2,4 mm thick discs.• Rest porosity about 15%• Permeancy measured of CO2 and He

Permeancy

1,50

2,50

3,50

4,50

5,50

1,50E-03 2,00E-03 2,50E-03 3,00E-03 3,50E-03

1/T

Ln(F

low

)

HeCO2

• No significant difference between He and CO2(Knudsen-like)

• At 250oC the material deteriorates and permeancy goes up

Materials Science-CO2 transport

Hydrotalcites• Hydrotalcites exist in a small compositional window

around Mg/Al=0.64/0.36=1.8• Hydrotalcites are not stable above 200oC

- dense HTC membranes are not feasible in the T,p window of the applications

- porous HTC membranes not first choice- search for alternative hydroxidic materials or porous supports

impregnated with hydroxidesCatalyst

• Four (pre)commercial pre-reforming catalysts have been tested.• Cheap Ni-catalyst is promising and could do the job in reformers with

CO2 selective membranes.

Materials Science-Conclusions

Reactor Design/System analysis• CO2 selective membranes show a too low CH4 conversion

- driving force for CO2 permeation to low to enhance reform reaction• CO2 selective membranes for WGS give CO conversions comparable to

H2 separating membranes and offer therefore a viable alternative- from efficiency point of view steam sweep should be as low as

possible

Materials Science• Hydrotalcites are not stable above 200oC

- Dense HTC membranes are not feasible in the T,p window of the foreseen applications

- porous HTC membranes not first choice- search for alternative hydroxidic materials or porous supports

impregnated with hydroxides

Summary