Chemical Looping Reforming with Packed Bed...

V. Spallina, B. Marinello, M.C. Romano, F.

Gallucci, M. van Sint Annaland

Chemical Looping Reforming with Packed Bed

Technology: Experimental Study and Modelling

Chemical Looping Reforming with Packed Bed Technology: Experimental Study and Modelling

2

OUTLINE

Motivation of the study

Description of the concept

Experimental study

Conclusions

Future works


3

Chemical Looping Reforming has been studied only using

interconnected Circulating Fluidized Beds but H2 production (as well as

CH3OH) is more convenient (techno-economic) at high pressure

→ Packed Bed Reactor

Fuel-to-chemicals conversion is less demanding in terms of heat

management than fuel-to-heat: the overall heat of reaction is lower

when compared to fuel combustion

→ CLR vs CLC

O2 (from ASU) and H2O (from steam turbine extraction) are not for free

→ auto-thermal reforming using oxygen carrier

CCUS applied to chemical industry

→ CO2 capture convenient without subsides (e.g. carbon tax)

Exploiting chemical looping technology in other industrially relevant

processes

→ chemical looping convenient without CO2 capture policies

Motivation of the study


4

the main concept

Air

N2

PSA offgas

CO2+H2O

CH4 (+H2O)

Reformed syngas

OX

IDA

TIO

N

RED

UC

TIO

N

REF

OR

MIN

G

Me (i.e. Ni)

MeO (i.e. NiO)


5

the concept for H2 production

OX

IDAT

ION

RED

UC

TIO

N

REF

OR

MIN

G

HT-WGS PSA

PSA offgas (CH4 2.5%, CO 13.7%, H2 28.3%, CO2 55.5%)

CO2 + H2O

H2

gas turbine

Air N2

steam

steam

ηref= 75.4% HR=3.04-3.14 Gcal/kNm3

H2

EE= 30-40 kWh/kgH2

deS HP-pure CO2 to CPU

H2O

H2O natural

gas

SH steam at (500˚C, 100 bar)

Export steam (180˚C, 6 bar)

Reformate (39.4%H2, 29%H2O, 19.2%,

CO, 11% CO2, 1.1% CH4)


6

units FTR-SMR FTR-SMR CLR-PBR

CO2 capture NO MDEA H2O cond

reforming efficiency (LHV basis) % 74 69 75.4

Heat Rate (no steam export) Gcalfuel/kNm3

H2 3.45 3.71 3.43-3.55

Heat Rate (with steam) Gcal/kNm3

H2 3.21 3.69 3.04-3.14

Electricity (prod/cons) GWh/kgH2 0.3 -10 30-41.8

Advantages: The system is convenient also when CO2 capture is considered Compared to FTR-SMR there are less conversion/separation steps Steam to process is significantly lower: S/C for FTR is 3.4; in this process is 1.9 (comb. ATR/FTR)

Conventional turbomachines are considered

The technology is flexible to different applications (i.e. Methanol, Fischer-Tropsch, etc..)


1D adiabatic axially dispersed packed bed reactor model: details are in

Spallina (2013, 2015), Gallucci (2015), Hamers (2013, 2014)

The reactor model has been validated for CLC using Cu, Ni, Fe and Mn

oxygen carriers

The model has been used for design and optimization of dynamically operated

PBR

Pseudo-homogeneous model

Radial temperature or concentration gradients are neglected

Heat losses through the reactor wall are neglected (not true for lab-scale

experimental validation)

Kinetics of Ni/CaAl2O4 G-S reactions and heterogeneous reactions are based

on Medrano(2015)

7

the model

1. Spallina V, Gallucci F, Romano MC, Chiesa P, Lozza G, van Sint Annaland M. Investigation of heat management for CLC of syngas in packed bed reactors. Chem Eng J 2013;225:174–191.

2. Hamers HP, Gallucci F, Cobden PD, Kimball E, van Sint Annaland M. A novel reactor configuration for packed bed chemical-looping combustion of syngas. Int J Greenh Gas Control 2013;16:1–12.

3. Gallucci F, Hamers HP, van Zanten M, van Sint Annaland M, Experimental demonstration of chemical-looping combustion of syngas in packed bed reactors with ilmenite, Chem Eng J 2015;274:156–168

4. Medrano JA, Hamers HP, Williams G, van Sint Annaland M, Gallucci F, NiO/CaAl2O4 as active oxygen carrier for low temperature chemical looping applications, accepted for publication (Applied Energy)


8

the model

REDUCTION REFORMING OXIDATION

Air

N2

PSA offgas

CO2+H2O

(CH4, H2O: S/C 2)

Reformed syngas

OX

IDA

TIO

N

RED

UC

TIO

N

REF

OR

MIN

G

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

700

750

800

850

900

950

1000

1050

0 30 60 90 120 150 180 210

gas f

racti

on

[v

ol.]

Tem

pera

ture

[°C

]

time cycle [s]

CO2

CH4

H2

CO

H2O

Temperature

H2O

Reduction with PSA off-gas leads to full gas conversion

and the gas is delivered at high temperature

The reforming step is providing H2-rich gas at the

equilibrium conditions. Due to the lower temperature, the

CH4 conversion decreases at the end of reforming.

During Oxidation the Gas temperature is in the range of

770-800˚C


9

the model

Air

N2

PSA offgas

CO2+H2O

(CH4, H2O: S/C 2)

Reformed syngas

OX

IDA

TIO

N

RED

UC

TIO

N

REF

OR

MIN

G

The OXIDATION step heats up the bed (850-900˚C) The REDUCTION with PSA-offgas moves the heat front

to the reactor outlet (cooling less than 30% of the bed) The REFORMING acts as heat removal:

the heat front cools down the reactor ‘from left to right’

the reaction front cools down the reactor ‘from top to down’

Reactor Inlet Reactor Outlet

550

600

650

700

750

800

850

900

950

1000

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Te

mp

era

ture

[°C

]

reactor length [m]

Solid Temperature Profile

After OX

After RED After REF


10

the experiments

small size PBR (heated) (about 500 g of OC/catalyst) 10-20 L/min flow rate (1 bar up to 1100˚C)

medium size PBR (semi-adiabatic) (about 5 kg of OC/catalyst)

50-160 L/min flow rate (7 bar up to 1100˚C)


11

the experiments

small size PBR (heated) 10 L/min flow rate (1 bar up to 1100 ˚C)

Insulation material

Vent

Mass spectrometerH2O

AirN2

H2

COCO2

Steam

oven oven oven

Packed bed reactor

PCWater condenser

T

48

T

1

p p

Reactor Length 0.5 m Ni-CaAl2O4 as catalyst/oxygen carrier

REACTIONS: RED: H2 + NiO → H2O + Ni (ΔHR = -15 kJ/moli) CO + NiO → CO2 + Ni (ΔHR = -47 kJ/moli) CH4 + 4NiO → 2H2O + CO2 + 4Ni (ΔHR = 134 kJ/moli) REF: CH4 + H2O ↔ CO + 3H2 ΔHR = 206 kJ/moli)

CH4 + CO2 ↔ 2CO + 3H2 ΔHR = 247 kJ/moli)

CO + H2O ↔ CO2 + H2 ΔHR = -41 kJ/moli)

OX: ½ O2 + Ni → NiO (ΔHR = -468 kJ/moli)


REDUCTION

12

The experiments

The stability of the OC has been tested for more than 200h of operation by checking the breakthrough of the H2 using 10 l/min (20 H2 / 20 H2O / 60 N2)

0

0.05

0.1

0.15

0.2

0.25

0.3

0:00:00 0:04:19 0:08:38 0:12:58 0:17:17 0:21:36

H2 m

ol f

ract

ion

, dry

time [ hh:mm:ss]

H2 breakthrough

H2_red#1 H2_ref#1

H2_ref#2 H2_ref#3

H2_ref#4 H2_aug24th

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0:00:00 0:01:44 0:03:27 0:05:11 0:06:55 0:08:38

gas

mo

l fra

ctio

n, d

ry

time [hh:mm:ss]

syngas breakthrough 10L/min (H2/H2O/CO2/N2 20/10/20/50)

CO2

H2

CO

Different breakthroughs occur in presence of syngas

(H2/CO2/H2O 20/20/10) Reverse WGS after the OC reduction

has been completed

1st test (OC not completely activated)

Reverse WGS


REFORMING

13

The experiments

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

0:00:00 0:00:43 0:01:26 0:02:10 0:02:53 0:03:36 0:04:19

Re

form

ate

mo

l fra

ctio

n, d

ry

time [hh:mm:ss]

Reformate composition 10 L/min (CH4/CO2/N2 10/60/30)

CO2

H2

CO

CH4

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.45

0.50

0:00:00 0:01:26 0:02:53 0:04:19 0:05:46 0:07:12

Re

form

ate

mo

l fra

ctio

n, d

ry

time [ hh:mm:ss]

Reformate composition 10 L/min CH4/H2O/N2 10/50/40)

CO2

H2

CO CH4

Different industrially relevant compositions (combination of CH4/H2O/CO2) diluted with N2 have been tested

The experiments have been done at 900˚C, the same tests will be carried at 800˚C

The CH4 full conversion has been achieved in both cases (working with high H2O or CO2 dilution)

The methane reforming (both steam and dry) reduces the temperature → carbon deposition control has to be taken into account

Steam Reforming

Dry Reforming

300

400

500

600

700

800

900

1000

0 0.1 0.2 0.3 0.4 0.5

T [°

C]

Reactor length [m[

Bed Temperature profile (Reforming)

REF_start REF 33%

REF 66% REF_end


OXIDATION

14

Experiments and Preliminary validation

0.00

0.05

0.10

0.15

0.20

0.25

0:00:00 0:01:26 0:02:53 0:04:19 0:05:46 0:07:12 0:08:38 0:10:05 0:11:31

O2 v

ol.

fra

ctio

n [

%]

cycle time [hh:mm:ss]

experiment model

the temperature rise is almost 600˚C) very close to the adiabatic value (≈700˚C) The thermal model is not able to catch the temperature rise at the beginning of the reactor (heat losses model need to be refined and adjusted)

the kinetic model is predicting the breakthrough of the oxygen with high accuracy → results confirms the previous experiments by Kooiman et al. (2015) using the same OC and reactor

300

400

500

600

700

800

900

1000

0 0.1 0.2 0.3 0.4 0.5

Tem

pe

ratu

re [

°C]

reactor length [m]

exp_30s exp_60s exp_90s

model_30s model_60s model_90s


Chemical Looping Reforming is a very promising way to combine H2 production and CO2 capture

A preliminary assessment shows already the gain (+1.5 percentage points compared to benchmark technologies)

Several industrial relevant processes can benefit from chemical looping

Commercial Ni-based OC has been (is being) successfully tested in a lab-scale reactor (more than 200 h)

The first lab tests confirm the proof-of-principle (at 900˚C and 1 bar for 10 L/min of syngas using different composition)

Carbon deposition play an important role due to the temperature variation (high syngas dilution)

Reformate composition is different than conventional plant (is that a problem in case of plant retrofitting?)

15

Conclusions

…but the way is long →


Experiments

Mapping the operating conditions of the reactors: composition,

temperature, dilution (i.e. S/C), flow rate

Oxygen carrier characterization and stability test (under multiple cycles)

Scale-up from 10 L/min (small PBR) to 100 L/min and 5 bar @TU/e

16

Future research

Modelling

Design of the reactor

Investigation with different reactor concepts

Heat transfer modelling (i.e. heat losses, etc…)

Heat management strategies

Techno-economic assessment of full scale plant and comparison with

conventional technologies for H2 production

Techno-economic assessment of the system for different process

(CH3OH, Fischer-Tropsch, etc..)


17

THANK YOU FOR YOUR ATTENTION!

CONTACT INFO: Vincenzo Spallina

mail: [email protected]

Chemical Looping Reforming with Packed Bed...

Documents

Transcript of Chemical Looping Reforming with Packed Bed...