New Developments in Combustion

Technology

Geo. A. Richards, Ph.D.

National Energy Technology Laboratory

U. S. Department of Energy

2014 Princeton-CEFRC Summer School On Combustion

Course Length: 6 hrs

June 23-24, 2014

‹#›

Energy - Everyday, Everywhere

Healthcare, education, infrastructure, water, transportation, communication, agriculture, recreation………

‹#›

Energy - Everyday, Everywhere….

….Except……

From the International Energy Agency Web-site:

“….Based on this updated analysis, we estimate that

in 2009 the number of people without access to

electricity was 1.3 billion or almost 20% of the

world’s population…..”

http://www.worldenergyoutlook.org/resources/energ

ydevelopment/accesstoelectricity/

‹#›

Gas 23%

Nuclear 10%

Renewables 25%

Oil 2%

Coal 40%

Gas 34%

Nuclear 16% Renewables

16%

Oil

‹#›

Why not my way?

“It’s all regional. It’s all local. And we just have to descend to that level to

judge it.” – Vaclav Smil, discussing preferable energy resources, Wall Street

Journal, Wed April 9, 2014, pp. R-1, Business and Environment special section.

Coal

Wind

Hydro

Gas

Nuclear

Solar

‹#›

Why not my way?

One size does not fit all

‹#›

Energy and Carbon Dioxide • Carbon dioxide capture and storage – costly, but not explicitly required.

• Carbon dioxide utilization in enhanced oil recovery (EOR) is needed, now.

• Carbon dioxide costs from natural source < anthropogenic sources.

Graphics and information NETL. Reference: DiPietro, J. P., Next generation Enhanced Oil Recovery, Presented at

the Carbon Dioxide Utilization Congress, San Diego, California, Feb. 19, 2014.

Available at http://netl.doe.gov/research/energy-analysis/publications/details?pub=68d576c3-a2ac-4e77-8963-

5ada3b04984f

Domestic EOR CO2 Use*

* A typical 550 MW coal plant emits 3.5 million tonne/ year

CO2;http://www.netl.doe.gov/energyanalyses/refshelf/PubDetails.a

spx?Action=View&PubId=348

Can we develop efficient &

affordable methods to supply CO2 ?

CO2 supply for North America EOR

million tonne/yr

Delivered CO2 prices today ~ $10-40/tonne

‹#›

Greenhouse Gas New Source Performance Standard (Proposed standard for new sources; comments accepted to May 9, 2014. Differs from proposed emission standards for existing

sources, released June 2, 2014, http://www2.epa.gov/carbon-pollution-standards)

0

500

1,000

1,500

2,000

2,500

Lb

CO

2/M

Wh

Existing

Subcritical PC

New

SC

PC NSPS

Limit

New

Uncontrolled

NGCC New

Supercritical PC

90% CCS New

NGCC

90% CCS

“Cost and Performance Baseline for Fossil Energy Plants, Volume 1: Bituminous Coal and Natural Gas to Electricity”,

Revision 2a, September 2013,

$70 $60

$21

$18

$32

$27

$0

$20

$40

$60

$80

$100

$120

$140

Subcritical PCRetrofit, 90% CCS

New Supercritical,1,100 Lb CO2/MWh

$/t

on

CO

2

First of a kind

Next of a kind (high range)

Next of a kind (low range)

$123/ton CO2

$105/ton CO2

Cost of CO2 Capture, $/ton CO2

“Cost and Performance Baseline for Fossil Energy Plants, Volume 1: Bituminous Coal and Natural Gas to Electricity”, Revision 2a, September 2013

‹#›

Carbon dioxide capture options

(1) Add a flue gas CO2

scrubber (e.g. photo below).

(2) Convert hydrocarbons to

hydrogen and CO2/w

separation/capture.

(3) Separate oxygen from air

and use oxy-fuel

combustion.

• Existing options: costly!

• Next generation options:

depend on thermal

science research:

– Supercritical carbon dioxide

power cycles: 300 bar

combustion?!

– Pressurized oxy-fuel.

– Chemical looping

combustion.

– Pressure-gain combustion

for efficiency.

– “Direct Power Extraction”

via MHD

240 MWe slipstream at NRG Energy’s

W.A. Parish power plant – Note the

size of CCS process area. Photo courtesy Mike Knaggs, NETL

‹#›

The role of capture AND generator efficiency

• A simple

heat/energy

balance defines

the overall

efficiency hov with

a carbon

separation unit.

• Reducing the

penalty from

carbon capture

comes from

BOTH:

– Decreasing wCO2

– Increasing hg

Generator Carbon

Separation

Unit

Q = mfDH Fuel Heat

Input

hg Generator

Efficiency

Wo Gross

Generator

Work

W1 Net

Output

Define:

a = (kg CO2 produced) / (kg fuel burned )

wCO2 = (separation work, Joules ) / (kg CO2) CO2

Approx Ranges: (30 – 60%) (6-10%)

‹#›

This presentation

Updated, expanded from 2012 CEFRC lecture:

– Inherent carbon capture: chemical looping combustion (Day 1)

– Step-change in generator efficiency: pressure gain combustion (Day 2)

– Frontier approach (?!): making oxy-fuel an efficiency advantage (Day 2)

P-gain rig @ NETL

RDC

Sampling

&

Diagnostics

Flow

‹#›

Disclaimer

* This report was prepared as an account of work sponsored by an agency of the United States

Government. Neither the United States Government nor any agency thereof, nor any of their employees,

makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy,

completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that

its use would not infringe privately owned rights. Reference herein to any specific commercial product,

process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute

or imply its endorsement, recommendation, or favoring by the United States Government or any agency

thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the

United States Government or any agency thereof.

*

‹#›

Chemical Looping Combustion

‹#›

Oxy-fuel background

Oxy- fuel achieves carbon capture very easily:

Air-Combustion:

CH + 5/4(O2 + 3.8N2) CO2 + 1/2H2O +4.7 N2

Oxy-Combustion:

CH + 5/4(O2) CO2 + 1/2H2O

“Usual” oxy-fuel approach: oxygen diluted with

CO2 or H2O added to an existing boiler cycle.

– Dilution used to keep the temperatures same as

existing cycle.

– Efficiency of the plant is penalized by the energy

needed to make oxygen.

Significant oxy-fuel demonstration projects are

occurring around the world.

– See for example: http://www.newcastle.edu.au/project/oxy-fuel-working-group/demonstrations.html

– More than 14 demos >10MWth listed

Costly to extract the CO2 from the N2 with amines

Easy to extract the CO2 from the H2O via condensation

Meridosia Illinois – Future Gen 2.0 planned site

Courtesy University of Utah – oxyfuel burner tests

http://www.newcastle.edu.au/project/oxy-fuel-working-group/demonstrations.htmlhttp://www.newcastle.edu.au/project/oxy-fuel-working-group/demonstrations.htmlhttp://www.newcastle.edu.au/project/oxy-fuel-working-group/demonstrations.htmlhttp://www.newcastle.edu.au/project/oxy-fuel-working-group/demonstrations.htmlhttp://www.newcastle.edu.au/project/oxy-fuel-working-group/demonstrations.htmlhttp://www.newcastle.edu.au/project/oxy-fuel-working-group/demonstrations.htmlhttp://www.newcastle.edu.au/project/oxy-fuel-working-group/demonstrations.html

‹#›

Making oxygen for oxy-fuel

• Oxygen can be supplied today by commercial Air Separation Units

(ASU) based on established cryogenic separation.

• The energy needed to separate oxygen from air is significant (see

below).

• In conventional oxy-combustion, we dilute the purified oxygen to

maintain the same boiler flame temperature as in air-combustion.

Air

Separation

Unit

(ASU)

1 mole of air

0.21 moles oxygen pO2

= 0.21 atm

0.79 moles nitrogen

pN2 = 0.79 atm

0.21 moles oxygen

pO2 = 1 atm

0.79 moles nitrogen pN2

= 1 atm

Reversible separation work:

~6 kJ/gmol O2 produced*

Current actual process:

~18kJ/gmol O2 produced**

*e.g, the change in gibbs energy for ideal mixing (Sandler, Chemical Engineering Thermodynamics (1989) pp. 313.

**See Trainier et al., “Air Separation Unit…..” Clearwater Coal Conference, 2010.

C + O2 CO2 DH ~ DG = 394 kJ/gmol (C or O2)

In efficient powerplants we convert

less than ½ of DH to work.

Thus~200kJ/gmol O2 work produced

Roughly ~1/10 of that is needed for ASU.

Dilute again

with CO2 or steam

‹#›

Chemical Looping • Shares advantages of oxy-fuel

– Product is just CO2 and H2O

• No separate oxygen production is needed

• Schemes for H2 production, carbon capture…

CO2 + H2O

Ash

Recycle

CO2 + H2OFuel

Air

Seal

Seal

N2 + O2

(vitiated air)

Carbon + metal oxide = CO2 + metal

Metal + air (oxygen) = metal oxide

CO2 + H2O

Avoiding confusion

with nomenclature:

Refer only to the

FUEL REACTOR

AIR REACTOR

Here’s why it can otherwise be confusing:

the air reactor “burns” or oxidizes the metal.

The fuel reactor “reduces” the metal oxide but

oxidizes the fuel.

Thus, you could call the fuel reactor an oxidizer for

the fuel OR a reducer for the metal oxide.

Today’s

discussion:

Focus on

air reactor

‹#›

Not quite new

• Chemical looping has been

around – but for different

reasons and applications.

– 1954 patent to manufacture CO2

• Similar process: iron-steam

route to hydrogen (circa 1920)*

Reduce iron with fuel and oxidize it with steam:

F3O4 + 4 CO 3Fe +4 CO2 3 Fe + 4 H2O Fe3O4+ 4H2

“Production of Pure Carbon Dioxide” US Patent 2,665,972 (1954)

Notice the heat exchangers (HX) in BOTH fuel and air reactors.

Should have made it a boiler?

Air Reactor

M + (O2 + 3.8N2)

MO2 + 3.8 N2 Fuel Reactor

MO2 + C

CO2 + M

M

MO2

CO2

HX

• And, before that….respiration.

Hemoglobin “loops”

to carry oxygen

from lungs for

hydrocarbon

oxidation in cells. *Hurst, S. (1939). “Production of Hydrogen by the Iron-Steam Method”, Journal of the American Oil Chemist’s Society, 16 (2), pp. 29-36.

‹#›

Basic Thermodynamics

In CL combustion, the overall reaction (1) of fuel with oxygen is split into two steps (2&3) , which add to the overall.

Consider an example of carbon and a metal/metal oxide (M/MO):

1) C + O2 CO2 DH1 Overall fuel oxidation - exothermic DH1 = DH2 + DH3 _______________________________________________________ 2) C+ MO2 CO2 + M DH2 Metal oxide reduction & fuel oxidation – can be endothermic OR exothermic

3) M + O2 MO2 DH3 Metal oxidation - exothermic

Nomenclature used in this talk:

Exothermic carriers Endothermic carriers Neutral carriers

DH20 DH2~0

Fuel reactor

releases heat Fuel reactor

consumes heat

Fuel reactor

does not consume or

release heat

Fuel reactor

Air Reactor

‹#›

Chemical Looping Heat Release

1) C + O2 CO2 DH1 Overall fuel oxidation - exothermic _______________________________________________________ (2) C+ MO2 CO2 + M DH2 Metal oxide reduction & fuel oxidation – can be endothermic OR exothermic (3) M + O2 MO2 DH3 Metal oxidation - exothermic

(3) (2)

CO2 + H2O

Ash

Recycle

CO2 + H2OFuel

Air

Seal

Seal

N2 + O2

(vitiated air)

CO2 + H2O

Air reactor: always exothermic; it releases heat.

Exothermic carriers: some heat will also come out in the

fuel reactor

Endothermic carriers: fuel reactor needs heat to reduce

the metal.

• If you don’t put heat into the fuel reactor, the

temperature will drop as reaction (2) proceeds, and the

reactions will stop.

• You add the heat by carrying it with the oxygen

carrier. Thus, the temperature of the carrier drops

some DT from inlet to exit of the fuel reactor.

• The heat flow rate is then (mass flow) x (CpDT ) and

must balance the heat used by reaction (2)

• Note that steam pipes won’t work to transfer heat into

the fuel reactor (>800C input, typically - exceeds

steam piping temperature limits).

‹#›

A useful concept: Chemical looping creates a “meta-fuel” for the air reactor

• A useful way to think about chemical looping combustion :

– The fuel reactor takes the hydrocarbon fuel (coal, oil, biomass,

natural gas) and uses that to make a different fuel (Cu, Fe, FeO,

Fe3O4, CaS…etc; the reduced oxygen carrier).

– For convenience, call the reduce oxygen carrier a “meta-fuel”*

*This is not a new phrase; a web-search reveals that the phrase was used to describe (inedible) tablets of fuel sold for camp stoves dating back to the 1920s:

“…..‘Meta Fuel’ is now extensively used to replace methylated spirit for such purposes as are fulfilled by small spirit lamps and stoves. It acts as an efficient substitute in such

circumstances, and has the advantage of being a solid substance, and thus easily portable and specially convenient. It is sold in small lamps and stoves, and refills are dispensed in the

form of white tablets or cakes. Judging from my experience in connection with the first case detailed below, it appears that many who sell this material are ignorant of its composition

and nature. Its poisonous properties on ingestion can hardly be too widely known……” R. Miller (1928). Archives of Diseases in Childhood. 3(18): 292–295.

Coal Iron oxide

“meta-fuel”

Burns w/o

any CO2.

Recyclable!

‹#›

Potential oxygen carriers

Many oxygen carriers have been studied to date: Iron: Fe2O3 Hematite = Iron (III), Fe3O4 Magnetite = Iron (II,III), FeO= Iron(II), Wusite, Fe Copper: CuO Copper oxide, Cu2O Cupric Oxide, Cu Nickel: NiO, Ni Manganese: MnO2, MnO, Mn2O3, Mn3O4 , Mn Cobalt: Co3O4, CoO, Co Sulfates-Sulfides: CaSO4-CaS, MnSO4-MnS, FeS-FeSO4 And others: Sb, Pb, Cd…

Fan, L. S., (2010). Chemical Looping Systems for Fossil Energy Conversions , John Wiley and Sons Publishers, see pp. 61 ff

Thermodynamics for iron and copper: Methane overall ½ CH4 + O2 ½ CO2 + H2O DH1000K = -402kJ Exothermic overall reaction Copper carrier 8 CuO + CH4 4Cu2O +CO2 + 2H2O DH1000C = -283kJ Exothermic metal reduction 4 CuO + CH4 4Cu +CO2 + 2H2O DH1000C = -211kJ Exothermic metal reduction 2 Cu + O2 2CuO DH1000K = -274kJ Exothermic metal oxidation Iron carrier 12Fe2O3 + CH4 8Fe3O4 + CO2 + 2H2O DH1000C = +154kJ Endothermic metal reduction 4Fe2O3 + CH4 8FeO + CO2 + 2H2O DH1000C = +303kJ Endothermic metal reduction 4/3Fe2O3 + CH4 8/3Fe + CO2 + 2H2O DH1000C = +154kJ Endothermic metal reduction 4/3Fe + O2 2/3 Fe2O3 DH1000C = -539kJ Exothermic metal oxidation

What large

difference in

system

configuration

must exist for

copper versus

iron carriers?

Hint: where does the heat

go, above?

‹#›

Thermodynamic limits on conversion

How much oxygen, CO, H2 will exist in the products of CL combustion? CO2 Enhanced Oil Recovery specification establishes potential requirements

*

*QUALITY GUIDELINES FOR ENERGY SYSTEM STUDIES -CO2 Impurity Design Parameters, DOE/NETL-341/011212, Jan 2012.

http://www.netl.doe.gov/energy-analyses/pubs/QGESSSec3.pdf

CO2 + H2O

Ash

Recycle

CO2 + H2OFuel

Air

Seal

Seal

N2 + O2

(vitiated air)

Notice there is no

“excess air” to

consume unused fuel.

‹#›

Understanding equilibrium limits on conversion

MeO2

Me

O2 (g)

The metal/oxide reaction (i) M + O2 MO2 ; DGT(i) = DGT(i)˚ + RTln(1/PO2) At equilibrium, DGT(i) = 0, denote PO2

* ; DGT(i)˚/(2.3RT)= log(PO2*) The gas-phase reactions (ii) 2CH4 + O2 2CO + 4 H2 ; DGT(ii)˚/(2.3RT)= 2log(PCH4

* / PCO*PH2

*2)+ log(PO2*) (iii) 2CO+O2 2CO2 ; DGT(iii)˚/(2.3RT)= 2log(PCO

* / PCO2*)+ log(PO2*)

(iv) 2H2+O2 2H2O ; DGT(iv)˚/(2.3RT)= 2log(PH2* / PH2O

*)+ log(PO2*) If you have the values for DGT˚’s you can solve immediately for PO2*, (PH2

* / PH2O*) and (PCO

* / PCO2*). You can get absolute concentrations of CO and H2

by noting the fuel is mostly converted to CO2 and H2O.

CH4, CO,

CO2, H2,

H2O

Notice that at any temperature

if PO2 < PO2* defined by (i), the

metal oxide (MO2)is reduced to

the Metal (M).

Quiz for grad students:

Your chemical looping

combustor is making 30

ppm CO.

Your professor wants you

to add more metal oxide

to improve CO burnout.

Will it work?

A) Yes because….

B) No because….

C) Maybe because….

D) I just want to

graduate.

‹#›

Figure shows the reduction of Fe2O3Fe3O4 with H2 or CO. Even at equilibrium, there are ppm levels of residual H2 & CO

making a slightly reducing environment. Note the combination of (ii) and (iii) is water gas shift CO + H2O CO2 +H2.

Residual CO increases with temperature because the CO Fe2O3 reduction reaction is slightly exothermic (H2 Fe2O3 reduction

here is endothermic). Results are based on simulations using HSC Chemistry 7.1. Courtesy Mike Gallagher, NETL.

Fe2O3 reduction to Fe3O4 with H2/CO The metal/oxide reaction Fe2O3 = hematite; higher oxidation, can reduce to magnetite.

(i) 6Fe2O3 O2 + 4Fe3O4

Fe3O4 = magnetite; lower oxidation, can oxidize to hematite, i.e., reverse of (i), or reduce further to FeO (Wustite), not discussed here.

If you fix temperature, (i) will proceed forward or backward depending on the oxygen pressure established with gas-phase fuel reactions, below:

The gas-phase reactions

(ii) 2CO+O2 2CO2

(iii) 2H2+O2 2H2O

‹#›

Fe2O3 reduction to Fe3O4 with H2/CO The metal/oxide reaction Fe2O3 = hematite; higher oxidation, can reduce to magnetite.

(i) 6Fe2O3 O2 + 4Fe3O4

Fe3O4 = magnetite; lower oxidation, can oxidize to hematite, i.e., reverse of (i), or reduce further to FeO (Wustite), not discussed here.

If you fix temperature, (i) will proceed forward or backward depending on the oxygen pressure established with gas-phase fuel reactions, below:

The gas-phase reactions

(ii) 2CO+O2 2CO2

(iii) 2H2+O2 2H2O

Notice what happens if you add (i) + (ii) :

2CO + 6Fe2O3 4Fe3O4 + 2 CO2

And similar for (i) + (iii):

2H2 + 6Fe2O3 4Fe3O4 + 2 H2O

The oxygen “disappears”.

Why bother referencing oxygen ?

As will be seen, it is easier to understand…

‹#›

Ellingham diagrams-1 These diagrams help quickly assess what species will

oxidize or reduce. Write any reaction with gaseous

oxygen as the reaction with one mole of O2, e.g.:

4Fe3O4 + O2 6Fe2O3 (rxn 1)

Then, at equilibrium, Drxn1G = 0 implies:

6GoFe2O3 -4GoFe3O4 – G

oO2 –RT ln (po2 ) = 0

ZERO

STD. GIBBS FREE ENERGY CHANGE FOR (rxn1), DGorxn1

LOOK UP IN TABLES, DEPENDS ONLY ON TEMPERATURE

Then, re-arranging the above equation

DGorxn1 = RT ln (po2 ) (eqn 1)

The equilibrium expressed by (eqn 1) is shown in the

graph as the intersection of the lines for different values

of oxygen partial pressure.

If the oxygen pressure is higher than equilibrium, (rxn 1)

will go forward. If it is less, (rxn 1) goes backward.

Fe2O3

Fe3O4

O2 (g)

CH4, CO,

CO2, H2,

H2O

What determines pO2? The gas reactions

DG

o r

xn

or

R

T l

n (

po2 )

k

J/gm

ol

O2

Temperature (K)

RT ln (po2 )

@ po2 = 10-8 atm

10-12 atm

10-16 atm

Note: You want it to go backward in the fuel reactor

‹#›

Ellingham diagrams-2 These diagrams help quickly assess what species will

oxidize or reduce. Write any reaction with gaseous

oxygen as the reaction with one mole of O2, e.g.:

4Fe3O4 + O2 6Fe2O3 (rxn 1)

Then, at equilibrium, Drxn1G = 0 implies:

6GoFe2O3 -4GoFe3O4 – G

oO2 –RT ln (po2 ) = 0

ZERO

STD. GIBBS FREE ENERGY CHANGE FOR (rxn1), DGorxn1

LOOK UP IN TABLES, DEPENDS ONLY ON TEMPERATURE

Then, re-arranging the above equation

DGorxn1 = RT ln (po2 ) (eqn 1)

The equilibrium expressed by (eqn 1) is shown in the

graph as the intersection of the lines for different values

of oxygen partial pressure.

If the oxygen pressure is higher than equilibrium, (rxn 1)

will go forward. If it is less, (rxn 1) goes backward.

Fe2O3

Fe3O4

O2 (g)

CH4, CO,

CO2, H2,

H2O

What determines pO2? The gas reactions

DG

o r

xn

or

R

T l

n (

po2 )

k

J/gm

ol

O2

Temperature (K)

RT ln (po2 )

@ po2 = 10-8 atm

10-12 atm

10-16 atm

What if you want to consider other

metal/oxide reactions

(e.g. Me + O2 MeO)?

Plot the DGo versus T for the metal of

interest; examples shown above (a)

and below (b) the Fe3O4 / Fe2O3 line.

b

a

‹#›

Ellingham diagrams - 3 4Fe3O4 + O2 6Fe2O3 (rxn 1)

Follow the exact same procedure to write the

Ellingham diagram for CO or H2

2CO + O2 2CO2 (rxn 2)

2H2 + O2 2H2O (rxn 3)

Consider (rxn 2); you can treat (rxn 3) in exactly the

same manner (not covered here). Then, at

equilibrium, Drxn2G = 0 implies:

2GoCO2 - 2GoCO - G

oO2 - RT ln (po2 pco

2 / pco22 ) = 0

DGorxn2 – 2RT ln(pco/pco2) = RT ln (po2 )

Notice you need to know the ratio of CO to CO2 partial

pressure to plot the left side versus temperature.

Assume some values for the ratio, and plot the left side

(10-2 , 3*10-5 and 10-7 ). The intersection with the

oxygen lines represents an equilibrium condition.

The plots match your chemical intuition – higher CO

levels have lower O2 at equilibrium

DG

o r

xn

or

R

T l

n (

po2 )

k

J/gm

ol

O2

Temperature (K)

RT ln (po2 )

@ po2 = 10-8 atm

10-12 atm

10-16 atm

10-16 atm

10-12 atm

RT ln (po2 )

@ po2 =

10-8 atm

500 1000 1500

Higher CO

Lower

O2

DG

o r

xn

or

R

T l

n (

po

2 )

k

J/gm

ol

O2

‹#›

Ellingham diagrams - 4 The last step is to add the equilibrium for the metal

and oxide reaction to the CO reaction, i.e. combine

rxn1 and rxn2 plots as shown in the bottom graph

4Fe3O4 + O2 6Fe2O3 (rxn 1)

2CO + O2 2CO2 (rxn 2)

The intersection of lines at (1000K, -225kj/gmol),

circled, represents equilibrium. • What is the oxygen partial pressure?

• The ratio of CO to CO2?

The graph provides insight into how changing

parameters affects the equilibrium of (rxn 1): • What happens if you raise the temperature?

• Add more CO?

• What happens if you consider a Me/MeO rxn that

“sits” above or below rxn 1 on the plot?

DG

o r

xn

or

R

T l

n (

po2 )

k

J/gm

ol

O2

Temperature (K)

RT ln (po2 )

@ po2 = 10-8 atm

10-12 atm

10-16 atm

10-16 atm

10-12 atm

500 1000 1500

P

10-12 atm

Lower

O2

Higher

CO/CO2

DG

o r

xn

or

R

T l

n (

po

2 )

k

J/gm

ol

O2

b

a

You can use published Ellingham diagrams with CO/CO2,

H2/H2O, O2 nomographs and various A +O2 AO2

‹#›

Quiz

• Your classmate wants to make an oxygen carrier from

aluminum because it is very energetic when it burns!

• How would you argue from experience if this is a good idea?

• How would you use an Ellingham diagram to figure out if that is

a good idea?

~1014

‹#›

Solid Carbon Formation • What happens if any solid carbon is left on the oxygen carrier

when it leaves the fuel reactor? (Red arrow, below)

• Carbon formation via equilibrium (chart, right) and also

hydrocarbon cracking.

• Notice that solid carbon on a metal oxide may not be a problem!

*Gaskell, D. R. (2008) Introduction to the Thermodynamics of Materials, 5th ed, Taylor and Francis, pp. 365-366

1.E-05

1.E-04

1.E-03

1.E-02

1.E-01

1.E+00

0 200 400 600 800 1000 1200

Vo

lum

e F

racti

on

CO

Temperature (C)

Boudard Reaction Equilibrium

C (s) +CO2(g) --> CO(g), 1atm, only CO/CO2 gases*

Carbon

Forms

Carbon

Gasified

to CO

CO2 + H2O

Ash

Recycle

CO2 + H2OFuel

Air

Seal

Seal

N2 + O2

(vitiated air)

‹#›

OXYGEN CARRIER CAPACITY AND

CIRCULATION RATES

Inert

support,

mass: minrt

Fully reduced

active species

(e.g., Cu, Fe, etc.)

Partially oxidized

active species

Fully oxidized

active species

(e.g., CuO, FeO, etc.)

active mass: mred

X = 0

active mass: mox

X=1

active mass: m

X = m-mred

mox-mred

Define

conversion

X for a

“supported”

metal oxide

carrier

‹#›

Mas

s of

carr

ier

(1 k

g o

xid

ized

sta

te)

Nomenclature

0.90

0.92

0.94

0.96

0.98

1.00

0 0.2 0.4 0.6 0.8 1

0.00

0.20

0.40

0.60

0.80

1.00

0 0.2 0.4 0.6 0.8 1

B

A C

C

DX X2 X1

A= Active mass, oxidized

B = Inert mass

C = Working oxygen capacity

co = A/(A+B)

Ro = C/A

co Ro = C/(A+B) co =

active

mass: m

mred

‹#›

Values for Oxygen Transport Capability (Ro) for Some Metal/Oxide Pairs

Table 1. Values of Ro for some potential oxygen carrier reactions, arranged small to large.

Inexpensive

Good capacity

Very inexpensive.

Throw-away option

Fe2O3 / Fe3O4 0.034

Mn2O3 / MnO 0.100

Mn2O3 / Mn3O4 0.034 Cu2O/ Cu 0.110

CuAl2O4 / CuAlO2 0.044 CuO / Cu 0.200

Fe2O3Al2O3 / FeAl2O4 0.045 CoO / Co 0.210

Co3O4 /CoO 0.067 NiO / Ni 0.210

Mn3O4 / MnO 0.070 Co3O4 /Co 0.270

CuAl2O4 /CuAl2O3 0.089 ZnSO4 / ZnS 0.396

NiAl2O4 / Ni Al2O3 0.091 CuS04 / CuS 0.401

CuO / Cu2O 0.100 MnSO4 / MnS 0.424

Fe2O3 / FeO 0.100 FeSO4 / FeS 0.425

CaSO4 / CaS 0.470

‹#›

Petrochemical

Fluid Catalytic Cracking (FCC)

Establishing Carrier Requirements

from FCC Experience Some FCC units operate with 41,000 kg/min solid circulation rate.

Proven Process Technology

Flue

Gas

Catalyst

Regenerator

Stripping

Stream

Air

Air Heater

Cyclone

Vessel

Dispersion Steam

Lower Feed Injection

Stripper

Stripper

Standpipe

Regenerator

Standpipe

Riser

Reactor

‹#›

Properties of the Oxygen Carrier

• Assuming a solids circulation rate ~ operating fluid

catalytic crackers (41,000 kg/min)

• Calculate the thermal output possible for a chemical

looping system for different carriers/conversions

• This only accounts for supplying oxygen, not thermal

balance (next slide).

0

200

400

600

800

1000

1200

1400

1600

1800

2000

0.5 0.7 0.9

Th

erm

al O

utp

ut

[MW

]

Conversion

[Pure Fe: Fe2O3→Fe3O4]

[Ilmenite: Fe2O3→Fe3O4]

[Ilmenite: Fe2O3→FeO]

[Cu, 60%Al: CuO→Cu]

[Pure Fe: Fe2O3→FeO]

[Pure Cu: CuO→Cu]

Acce

pta

ble

Op

era

ting

Ra

ng

e?

0 500 1000 1500 2000 2500

1

2

3

4

5

6

Thermal Output [MW]

[Pure Fe: Fe2O3→Fe3O4]

[Pure Fe: Fe2O3→FeO]

[Ilmenite: Fe2O3→Fe3O4]

[Ilmenite: Fe2O3→FeO] [Pure Cu: CuO→Cu]

[40% Cu, 60%Al2O3: CuO→Cu]

‹#›

Determining the solids circulation rate

• The oxygen carrier circulation rate is determined by:

1) The oxygen needed for the fuel flow rate

2) Endothermic carriers: the heat needed to drive the fuel reactor

• How are requirements (1) and (2) compatible?

– It turns out the required circulation rate is sometimes entirely

dominated by the need to supply heat to the fuel reactor.

– In that case, a higher capacity oxygen carrier is not needed.

‹#›

Circulation rate analysis

• Solids circulation shown

by the double curved

arrows, A and B.

• There is only ONE

circulation rate – why

show as two arrows?

– Emphasize that you

must satisfy two

conditions:

A. Supply enough oxygen to

react with all the fuel.

B. Endothermic carriers:

supply enough heat to

keep the carrier “hot”.

• Note that the mass flows

that satisfy A has a

definite minimum for a

fixed fuel flow rate, but

no maximum.

CO2 + H2O

Ash

Recycle

CO2 + H2OFuel

Air

Seal

Seal

N2 + O2

(vitiated air)

CO2 + H2O

Fuel flow

for 1MWth

(assume CH4)

A

B

CO2 H2O

Air in

Vitiated Air Out

Heat

Out

(steam)

Which is bigger? A or B ?

‹#›

Circulation rate analysis, continued

A – the mass flow needed to supply

enough oxygen to react with the fuel.

B- the mass flow needed to carry enough

heat to support the endothermic reaction

via a 50°C temperature drop.

A

B

CO2 H2O

Air in

Vitiated Air Out

Heat

Out

(steam) Fuel flow

for 1MWth

Fuel

Reactor Air

Reactor

A B B/A

Carrier Pair Ro Min Circ Rate O2 supply

(kg/sec)

Circ rate to supply

heat (kg/sec)

Ratio: Heat/O2

Rate

Fe2O3 / Fe3O4 0.034 2.35 3.36 1.4

Fe2O3 / FeO 0.1 0.80 6.75 8.4

Fe2O3/Fe 0.3 0.27 6.10 22.9

CuO / Cu 0.2 0.40 n/a n/a

CaSO4 / CaS 0.47 0.17 3.07 18.1

DT = - 50°C

Analysis is for methane fuel; could be repeated for coal, etc. Note that the exothermic carrier CuO has a much lower circulation rate than all the others.

‹#›

KINETIC RATES AND REACTOR SIZE

‹#›

KINETIC RATES AND REACTOR SIZE

Air

Fluid Bed

Carrier

State X1

Carrier

State X2

Inputs

moc,iXi mair,i

Ti

Reactor

Volume

VR

Outputs

moc,oX

mair,o T

Reactor

Temperature

T

A “bubbling” fluid bed (BFB) is one possible reactor configuration (left).

BFB is approximately like a stirred reactor (right).

The usual combustor design concepts apply: 1) heat release rate balances the incoming rate of cold reactants or it will blow out

2) for a given throughput, the reactor volume is inversely proportional to the reaction rate.

‹#›

thermocouple

Experimental measurement of kinetics Thermo gravimetric Analysis (TGA)

For gaseous reactions, sample weight describes the conversion:

Cycling the fuel and oxidizer over the sample pan will give

many “cycles” to analyze.

Solid fuels (coal and biomass): must re-load the pan every cycle.

‹#›

Example of kinetics measurement (CH4)

Typical mass and temperature measurement for CuO/bentonite

particle and 100% CH4 for reduction and air for oxidation

reactions

Effect of reaction temperature on CuO/bentonite

particle and CH4 reaction

0

0.2

0.4

0.6

0.8

1

0 0.4 0.8 1.2 1.6 2

Conv

ersi

on (

X)

Time (min)

750

800

850

900

T (oC)

0

200

400

600

800

1000

30

32

34

36

38

40

42

44

0 500 1000 1500 2000

Tem

per

atu

re (

oC

)

Mas

s (m

g)

Time (min)

1 rrox r

m mX

m m

=

dp=150-250 mm

TGA Data for Cu-based carrier 4 2 24 4 2CuO CH Cu CO H O

100% CH4, 800 oC

100% CH4

Monazam et al. (2012) “Kinetics of the Reduction of CuO/Bentonite by Methane (CH4) During Chemical Looping Combustion”, to appear Energy and Fuels.

4

11

(1 )[ ln(1 )]m nCHdX

ky n X Xdt

= Results fit to Jonson-Mehl-Avarmi (JMA) rate equation

‹#›

A visual representation of reaction rates

• An informative/interesting way to see the metal-metal oxide cycle.

• Combustion Quiz

– What type of flame does a propane torch use (e.g., diffusion, premixed,

partially premixed?)

– What is the partial pressure of oxygen inside the flame?

Movie of copper reduction and

oxidation

‹#›

Kinetics in a stirred reactor bed (1/2)

• With kinetic rates, write

energy and mass balances

with an “efficiency” of

conversion hI (defined below).

• Does the bed have “light-off”

behavior?

Green line depicts efficiency as a function of output

temp (from mass balance using kinetic rates), dashed

line shows same using energy balance equations.

Point of intersection is the desired steady-state

solution.

0

200

400

600

800

1000

1200

1400

1600

1800

2000

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Rea

cto

r In

tern

al T

emp

erat

ure

(C)

Efficiency (N-I)

Air Reactor (Cu2O-->CuO)

10 sec

800C - Input Temp

Tout at steady-

state

‹#›

Kinetics in a stirred reactor bed (2/2)

• With kinetic rates, write

energy and mass balances

with an “efficiency” of

conversion hI (defined below).

• Does the bed have “light-off”

behavior?

0

200

400

600

800

1000

1200

1400

1600

1800

2000

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Rea

cto

r In

tern

al T

emp

erat

ure

(C)

Efficiency (N-I)

Air Reactor (Cu2O-->CuO)

10 sec

800C - Input Temp

Tout at steady-

state

Different inlet

temperature, mass

flow, etc.

Green line depicts efficiency as a function of output temp

(from mass balance using kinetic rates), dashed line

shows same using energy balance equations. Point of

intersection is the desired steady-state solution.

Detailed approach: kinetics inserted

into CFD models (described later).

‹#›

How quickly do the reactions need to occur?

Answer: it depends on several parameters Simple estimate:

The extent of conversion is X.

Assume constant rate (dX/dt)

Residence time is t.

t (dX/dt) = DX (2) Assume we want DX ~ 1

and combine (1) and (2):

m = mass flow rate

“active” species*

M = “active” mass , raV

t = M/m, res. time

DH = heat release/mass

m DH Thermal output

V Volume =

ram DH

raV = ra DH / t

or, multiplying by ra/ra:

(1)

Thermal output

Volume =

(3)

raDH (dX/dT)

*In fluid bed boilers, the solids flow will include significant amounts of inert material like ash , limestone, sand .

[1] Steam, 41st edition, Babcock and Wilcox, see Figures 4, 12 in chapter 17 for bed density; (2% )discussed later.

Using bubbling fluid bed boilers as an example, you

can “check” this with reasonable parameters:

ra = (2%) rbed; rbed ~ 500kg/m3 [1]

t ~ 60 sec, Basu pp. 123

DH = 23MJ/kg, coal HHV Leads to volume heat release 3.8MJ/m3; “reasonable” see

Basu pp. 236

Note: (1) you can adjust ra

(2) t is tens of seconds, or more

‹#›

Solid fuel combustion

‹#›

TGA Profile of Coal +CuO in N2

Can We Use Coal Directly?

0 50 100 150 200 250

0

5

10

15

20

25

30

35

40

Reaction time (min)

weig

ht

(mg

)

0

200

400

600

800

1000

Reactio

n te

mp

era

ture

(oC

)

44.88%

55.12%

N2 O2

TGA Profile of Coal in N2

Volatiles &

mositure out

combustion

combustion

0 20 40 60 80 100 120 140 160 180

160

170

180

190

200

210

Reaction Temperature (oC)

Weig

ht

(mg

)

200

400

600

800

1000

Tem

pera

ture

(oC

)

Nitrogen Oxygen

0 20 40 60 80 100 120 140 160 180

160

170

180

190

200

210

Reaction Temperature (oC)

Weig

ht

(mg

)

200

400

600

800

1000

Tem

pera

ture

(oC

)

Nitrogen Oxygen

Reaction time (min)

Rates were higher than expected.

Why?

‹#›

Reaction Pathways for Solid Fuel CLC

• Coal CLC with metal oxides via gaseous intermediates:

In N2:

Coal Coal pyrolysis

CO/H2 + CuO Cu +CO2 /H2O

CO2 + C 2CO

In CO2:

C+CO2 2CO

CO+CuO Cu+CO2

• CLOU mechanism (Chemical Looping Oxygen Uncoupled):

CuO Cu/Cu2O +O2

Coal + O2 CO2

• Solid-solid interaction: MeO+C MeO +CO2

Discussed

next

‹#›

Possible Reasons for Rapid, Low-Temperature reaction

of solid carbon with CuO

2CuO = Cu2O + 1/2O2

At 500 oC, PO2 is 1.1*10-9

Will removal of oxygen (1.1*10-9)

continuously by carbon, facilitate the

CuO decomposition?

• Reacted C with various oxygen partial

pressures

– With air: modest reaction at 500-600oC.

– No reaction at 500-600 oC with oxygen at

low O2 partial pressure ( At 500oC

Significant

Reaction

at 500oC vitiated air

‹#›

Combustion Rates of Coal (100 micron) with

Various Particle Sizes of CuO in TGA

Higher combustion temperature with increasing particle size

0 20 40 60 80 100

0.00

0.02

0.04

0.06

0.08

0.10

Re

actio

n te

mp

era

ture

(oC

)~5 micron

713 oC

63-177 micron

780 oC

354-595 micron

874 oC

Re

actio

n r

ate

(m

in-1)

Reaction time (min)

0

200

400

600

800

1000

‹#›

Effect of Dilution by Quartz Powder on the TGA

Combustion Performance of Carbon and CuO

• Mixed CuO, Quartz and C

powders

• CuO/C ratio was kept

constant

• Reaction T increased with

increased dilution

These data and others

(flow tests + DFT

calculations)* suggest a

solid-phase reaction

between carbon and the

oxygen carrier (Fe, too).

Reduces coking automatically!

* Siriwardane, R. Tian, H., .Miller, D., Richards, G., Simonyi, T., Poston, J. (2010). Evaluation of reaction mechanism of coal-metal oxide interactions in

chemical-looping combustion, Combustion and Flame, Combustion and Flame, Volume 157, Issue 11, November 2010, Pages 2198-2208

‹#›

Reactor and System Design

‹#›

An introduction to fluidization

• Fluidization is used

widely in the

chemical industry – Fluid Catalytic Cracking

(Hydrocarbons)

– Catalytic reactions.

– Drying and calcining.

– Many reactions.

• Coal, biomass, and

waste combustion

or gasification.

• Chemical looping.

Main topics

• Physical description of

fluidization.

• Key velocity parameters:

minimum fluidization,

bubbling, and terminal.

• Fluidization regimes,

application, and

components.

• Effect of particle

morphology

• Reactor configurations.

‹#›

References

Basu, P.(2006), Combustion and Gasification in Fluidized Beds , CRC Press,

Taylor and Francis Group, Boca Raton, FL.

Excellent treatment of practical issues in design of fluid bed combustors and

gasifiers.

Kunii D., Levenspiel, O. (1991). Fluidization Engineering, 2nd ed.,

Butterworth-Heinemann, Newton MA.

A classic text on fluidization – get a copy if you work on multi-phase flows.

‹#›

Physical Description of Fluidization

• A granular material

does not typically

flow like a fluid; it can

form a “pile”; (a).

• But, if you supply

fluidizing gas,

granular material can

behave as a liquid:

– Horizontal surface (b)

– Flow from holes (c)

– Equalizes levels (d)

– Floats light objects (next slide)

gas

gas

mesh

No gas

(a) (b)

(c) (d)

No gas gas

‹#›

Physical description of fluidized beds (cont.) – Lighter objects float.

– The bed volume is larger in a

fluid state:

void fraction e = Vgas/(Vgas + Vsolid)*

– The gas flow rate is typically

described by the superficial

velocity* U: (Gas volume flow rate)

(Cross-section area, no solids present)

– The bed pressure drop

balances the overhead weight

gas

mesh

No gas

* Typical e values( - ), U values [ m/s]: Packed Bed (0.4-0.5) [1-3], Bubbling Bed (0.5-0.85)[0.5-2.5], Circulating Bed (0.85-0.99) [4-6], Transport Reactor(0.98-0.998)[15-30], Basu pp. 22

Larger

Volume

DP Axc = Axc Lb (1-emf) (rsolid-rgas)

DP

Lb

The mf subscript: voidage at minimum fluidization, explained next

gas

Floats!

Sinks!

‹#›

Velocity parameters

• Minimum Fluidization Velocity Umf:

– The superficial velocity that “just” fluidizes;

the point where the bed weight balances the

pressure drop.

– Typically you measure Umf to get an

accurate value for a new granular material.

• The minimum bubbling velocity Umb:

– The velocity where bubbles first appear.

– Can be equal or greater than Umf.

• The particle terminal velocity Ut:

– Note the velocity is NOT a uniform profile.

– What happens if U> Ut ?

gas

gas

gas

Shown smaller

to with more height

above the bed

Fluidization regimes: smooth, bubbling, turbulent,

fast, pneumatic transport

‹#›

Application to

chemical looping

• Air reactor (right), fuel

reactor (left).

• Solids are transported from

the air reactor where U>Ut.

• Solids are fluidized in the

fuel reactor where Umf

‹#›

Loop seal and L-valve components

Loop seal – Isolates process gas

above (A) from process gas below (B). L-Valve- controls the solid flow

delivered to the right.

Fluidizing gas

Fluidizing gas

(A)

(B)

Solid flow

‹#›

The particle properties: effect on fluidization Fluidization regimes depend on the

particle morphology, size, density.

Geldart Classification A, B, C, D

A = Aeratable. Can achieve smooth

fluidization, low density (

‹#›

Types of gas-solid reactors

• Fixed bed: Not fluidized, UUtr, the particles are carried

out of the bed, and are re-cycled.

Ex: CFB Boiler, FCC for

hydrocarbon cracking.

• Moving bed: Not neccesarily

fluidized, but the solid moves

countercurrent to the process

gas. Ex: Lurgi gasifier.

• Entrained: no “bed”, dilute

phase. Ex: Pulverized coal boiler

Circulating fluid bed combustor

‹#›

Discussion-

• The preceding fluidization introduction has covered

just hydrodynamics. Also need to consider reactions,

conversion, heat transfer.

• If you want to design a solid fuel chemical looping

boiler, what combination of gas-solid reactors is the

best?

• What are the tradeoffs for small or large oxygen

carrier particles?

• Some of these issues are best addressed with

validated CFD simulations and system models.

‹#›

Modeling of Fluidized Beds

(Courtesy: F. Shaffer, NETL)

DEM LBM DNS MP-PIC Multi-Fluids Filtered-Eqs ROM

www.mfix.netl.doe.gov

Movie

http://www.mfix.netl.doe.gov/

‹#›

Comparison of CFD and Cold Flow Rig

Oxygen

carrier

Solid fuel

into this bed

Lighter ash

carried out with

fluidizing steam

or CO2

Oxygen

carrier

SIMULATION EXPERIMENT

Air

reactor

Movie

‹#›

Design consideration – the air reactor

(fuel reactor discussed later)

‹#›

Where do you take the heat out?

Endothermic carriers: you don’t need need

to remove heat from the fuel side.*

Exothermic carriers, you may need to

remove heat from the fuel side to keep from

overheating the carrier. Potentially use the

recycle fluidizing gas as a heat exchange

media.

All carriers: must manage the air reactor

exotherm.

CO2 H2O

Air in

Vitiated Air Out

Heat

Out

(steam) Fuel flow

for 1MWth

Fuel

Reactor Air

Reactor

Endo or

Exo-

thermic?

‹#›

(Cont) where do you take the heat out – endothermic carrier?

Must allow equal temperature rise in the air reactor DTar as

temperature drop in the fuel reactor DTfr

CO2 H2O

Air in

Vitiated Air Out

Fuel flow

for 1MWth

Fuel

Reactor Air

Reactor

DTfr = - 50°C DTar = + 50°C

Heat

Out

(steam)

A simple enthalpy balance will

show that the air needed to

transport the carrier has a minor

effect on the carrier temperature -

even w/o exothermic reactions.

You will need to take a lot of heat

out as the reactions occur. How?

‹#›

Conventional Circulating Fluid Bed Combustion (CFB)

• Fuel is added to the “riser” and reacts with

air.

• Unburned fuel may

circulate around the

“loop” several times.

• The circulating “bed” of

material is mostly inert,

and provides a large

thermal mass.*

• Heat can be removed in

several places.

• Note the similarity to

chemical looping

combustion.

* “…..the mass flow rate of recycled solids is many times the mass flow rate of incoming air, fuel, and limestone….the bed solid temperature

remains relatively uniform” Steam, Edition 41, pp. 17-9, The Babcock and Wilcox Company.

‹#›

Circulating Fluid Bed (CFB) Combustion

versus Chemical Looping Air Reactor

• The CFB case uses

significant inert flow

that moderates the

temperature rise.

• The CLC case

appears more like

pulverized coal –

potential for

significant

temperature rise?

• Compare reaction

enthalpies (i.e., HHV)

for meta-fuel versus

carbon (next slide).

Solid

Fuel

Particles

Inert bed

Particles

(ash, sand)

Meta

Fuel

Particles

CFB

Combustion

CLC

Air reactor

Air Air

‹#›

Circulating Fluid Bed (CFB) Combustion versus Chemical Looping

Solid

Fuel

Particles

Inert bed

Particles

(ash, sand)

Meta

Fuel

Particles

CFB

Combustion

CLC

Air reactor

Air Air

Reactions 1 – 7 compare DH for meta-fuel (CLC) and

carbon fuel (CFB).

Per mass basis because CFB circulation is

expressed as mass flux (kg/s/m2).

The meta-fuel particles produce 3-60 times less heat

per mass than carbon fuel.

But, there can be significantly more mass flux of

meta- fuel versus carbon fuel (how much more?).

"meta" fuel or carbon fuel Reaction

Reaction DH per kg

'fuel', MJ/kg *

Iron carriers

1 Fe3O4 + 1/4O2 --> 3/2 Fe2O3 -0.51

2 FeO+1/4O2-->1/2Fe2O3 -1.96

3 Fe+3/4O2 --> 1/2Fe2O3 -7.39

Copper carriers 4 Cu+1/4O2 --> 1/2Cu2O -1.34

5 Cu+1/2O2 --> CuO -2.46

Calcium sulphide carrier 6 CaS + 2O2 --> CaSO4 -13.24

Carbon fuel 7 C+O2 --> CO2 -32.76

0

5

10

15

20

25

30

35

1 2 3 4 5 6 7

Reaction

DH

, J/k

g

* Standard reaction enthalpy at 298K; data calculated from NIST web book and Lange’s Handbook of Chemistry

Iron Copper

CaS

Carbon

‹#›

CFB Typical Flow Parameters

Parameter* Boiler FCC Reactor

External Circulation Rate (kg/m2/s) 10-50 500-1000

Superficial Gas Velocity (m/s) < 6 < 25

Suspension Density, upper region (kg/m3) 1-10 10-100

Particle Size (microns) 200 70

• The chart below provides typical parameters for

CFB operation in boiler and Fluidized Catalytic

Cracking applications.

• Using the superficial gas velocity for the boiler ~6

m/s and multiplying by the air density (800°C, 1

atm) provides a mass flux of air : Ga= 2 kg/m2/s.

• With the smallest circulation rate, from chart below,

Gs = 10 kg/m2/s. Solids to air mass ratio is 10/2 = 5.

*Basu, P.(2006), Combustion and Gasification in Fluidized Beds , pp. 42, CRC Press, Taylor and Francis Group, Boca Raton, FL.

** This ratio can cover a wide range of values; this value uses the parameters indicated.

Solid

Fuel

Particles

Inert bed

Particles

(ash, sand)

Air Mass Flux Ga

Solid Mass Flux Gs

Thus, Gs/Ga ~ 5 typical** solid/air mass ratio

‹#›

The relation between fuel/air ratio and Gs/Ga If the solids flow is meta-fuel (or, pure solid fuel):

fuel/air mass ratio = Gs/Ga

If the solids flow has a mass fraction of fuel Yf:

Yf = (fuel mass flux)/(solid mass flux)

fuel/air mass ratio = Yf ·Gs/Ga

The chart below lists the stoichiometric fuel-air

ratios for several meta-fuels and carbon fuel:

Most of the f/a ratios are less than Gs/Ga = 5

(except reaction 1).

Thus, you need to operate very dilute OR don’t

convert much of the material.

Coal CFBs accomplish dilute fuel operation with

the addition of lots of inert materials (~ 3/2 Fe2O3 + 0.94N2 6.73

2 FeO+¼(O2+3.76N2) -->1/2Fe2O3 + + 0.94N2 2.09

3 Fe+3/4(O2+3.76N2) --> 1/2Fe2O3 +2.28N2 0.542

Copper carriers 4 Cu+1/4(O2+3.76N2) --> 1/2Cu2O + 0.94N2 1.85

5 Cu+1/2(O2+3.76N2) --> CuO + 1.88N2 0.926

Calcium sulphide carrier

6 CaS + 2(O2 +3.76N2) --> CaSO4 + 7.52N2 0.266

Carbon fuel 7 C+(O2 +3.76N2) --> CO2 +3.76N2 0.0874

* For carbon fuel example, f/a= Yf·Gs/Ga is 0.0874 = Yf ·5. Thus, for stoichiometric conditions, Yf = .0175. Actual plants operate with excess air, lowering this value further.

‹#›

Practical issues

The air reactor residence time is not necessarily

long enough to allow full oxidation. • CFB riser residence time ~ 5 seconds, max.

• Full oxidation takes ~ >>10 seconds for some carriers.

Options: A. Supply approximately 10/5 = ~ 2x more carrier than

stoichiometric; convert part of the carrier (figure A, left) and use

external heat exchanger.

B. Add a bubbling bed at the bottom with in-bed heat removal

(figure B, below).

C. Arrange two interconnected loops (figure C, below):

D. Your idea? How about Moving Bed?

CLC

Air reactor

Air, Inlet

Fe3O4 or FeO

+ Fe2O3

Outlet

Fe3O4 or FeO

Fuel reactor

carrier circulation

Air reactor

carrier circulation

Exchange between reactors

Secondary

Transport Air

Steam

Generation

Tbed

Texit

Air A B C

External HX

External HX

‹#›

Advantages of case C (last slide)

Follow the mole flow rate of reduced species (subscript r) at

stations 1 – 5 shown in the diagram. Some definitions:

F = single pass fractional conversion; Nr3=(1-F)Nr2.

R = recycle rate; R = Nr5/Nr4 (see note, below)*.

F = 0 implies no oxidation occurs in the reactor.

F = 1 implies no reduced material emerges at station 3 – full oxidation.

R = 2 (for example) means the molar flow in recycle is twice the outflow.

Fuel reactor

carrier circulation

Air reactor

carrier circulation

Nr1

Nr4

Nr1 = moles/sec reduced state input

Nr4 = moles/sec reduced state output

Nr1

Nr4

F R

1

2

3

4

5

* Note that since the flow splits at 3 to 4-5, the recycle/output mole flow ratio is the same for reduced or total

molar flow (oxidized + reduced + inert, tot subscript); Nr5/Nr4 = (Yr5*Ntot5)/(Yr4*Ntot4). The mole fractions Yr5

= Yr4 because they are split from the same stream at station 3. Thus, Nr5/Nr4 = Ntot5/Ntot4 = R.

Nr4 = (1-F)

(1-RF) Nr1 Mole balances imply:

Air reactor process flow Equivalent topology & nomenclature

0

0.2

0.4

0.6

0.8

1

0 0.2 0.4 0.6 0.8 1

Nr4

/ N

r1R

ed

ue

d s

tate

flo

w o

ut/

in

F Single pass fractional conversion

Recycle

RR=0

R=1

R=2

R=4

R=8

Main point: Modest single-pass fractional

conversion can be offset by recirculation; e.g.:

F = 0.2, R = 8 implies Nr4/Nr1 = 0.3

( Only 30% is not reduced, 70% is oxidized)

Disadvantage: large R promotes attrition.

‹#›

Recent Research Results

Chemical Looping – work in progress at NETL

Cold flow validation rig, 50kWth Chemical Looping Reactor, Single fluid bed, attrition test, microwave solids flow sensor.

– Experiments: Designed, built, and producing validation data.

• Lab-scale to fully-integrated 50kWth loop.

– Simulations: Developed models, simulation tools . • Zero-D to full-loop 3-D.

– Developed novel performance/low-cost carriers. • ”Promoted” iron-ore, and “thermally neutral” Cu-Fe.

– Techno-economic results: results for NG steam. • Confirms low-cost & guides research (shown later).

Fuel Reactor

Separation cyclone

Riser

Crossover

Air Reactor

Loop Seal

L-Valve

Solids Flow

Models: zero-D, 3-D loop, fuel reactor

NETL O2 Carriers Maximal O2 capacity. “Thermally neutral”.

Std. materials & manufacturing

‹#›

Key Findings- Chemical Looping – A few more details

• Thermally neutral carrier development:

– Allows independent control of circulation rate.

– Large O2 capacity means low circulation rate

– reduced attrition?

• Hydrodynamic predictions versus

measurements:

– Cold flow predictions significantly affected by

chosen model parameters.

– Emphasizes need for model validation and

diagnostics in hot flow.

– Microwave sensor being developed for hot

flow diagnostic.

• Lessons learned and progress made

• Multiple solids flow issues resolved –

see (1) and (2) - recent test had smooth

solids circulation.

• Batch mode oxidation/reduction on

baseline (raw) hematite; follows

expected model results.

• Bubbles matter! See next slide

Cyclone

C-1200

Test Section

C-1250

Loop Seal

R-1300

Upper

Riser

R-1150

Lower

Riser

R-1100

Air

Reactor

R-1000

Air Pre-heater and Tee

H-1800 & H-1850Air Pre-heater and Tee

H-1800 & H-1850

L-Valve

Housing

R-1450

Fuel

Reactor

R-1400

(1)

(2)

Industrial Carbon Management Initiative

‹#›

Cyclone

C-1200

Test Section

C-1250

Loop Seal

R-1300

Upper

Riser

R-1150

Lower

Riser

R-1100

Air

Reactor

R-1000

Air Pre-heater and Tee

H-1800 & H-1850Air Pre-heater and Tee

H-1800 & H-1850

L-Valve

Housing

R-1450

Fuel

Reactor

R-1400

a a’ a a’

Movie (cold flow) a-a cut

Fuel Conversion and Bubbles

Porous plate different injectors Video and CFD courtesy Doug Straub, A. Konan, NETL

‹#›

Fuel conversion: CFB versus chemical looping fuel reactor

CO2 + H2O

Ash

Recycle

CO2 + H2OFuel

Air

Seal

Seal

N2 + O2

(vitiated air)

Circulating fluid bed combustor

• Gas-phase combustion for CO &

H2 has ~ 3% oxygen.

• Solid char is recycled.

• Gas-phase combustion for CO & H2

requires oxygen carrier (gaseous

oxygen PO2

Techno Economic Studies : Parameter Potential Impacts Expected

Relative to a CLC Base case (assumes changes in parameters are within operating range)

Vessel Height

Vessel Diameter

Circ. Rate

Boiler Eff. Auxiliary

Power CO2

Capture Equip. Cost

Cost of Steam

Carrier Reactivity (literature)

Large + = -

Small + = -

Small + = -

Small + = -

Carrier Loss (0 %) and Price ($0/lb)

Medium + = +

Carrier Size (0.28mm) and Density (203 lb/ft3)

Small + = -

Small + = -

Small + = -

Small + = -

Carrier Conversion (from reducer 47%; from oxidizer 95%)

Medium + = +

Large + = -

Small + = +

Small + = +

Small + = +

Reactor Temperature (1700 F)

Small + = -

Small + = +

Small + = -

Small + = -

Small + =-

Reactor Velocities (reducer outlet 33.6 fps; oxidizer outlet 29.4 fps)

Large + = +

Large + = -

Small + = +

Small + = +

Small + = +

Natural Gas Conversion (97.5%)

Medium + = +

Small + = +

Small + = +

Large + = +

Small + = +

Small + = +

Oxidizer XS O2 (3.8mol % in off-gas)

Small + = -

Small + = +

Small + = -

Small + = +

Small + = +

Small + = +

This is why we want to focus on this issue going forward

Industrial Carbon Management Initiative

‹#›

Chemical Looping – Discussion/ Thinking Question

• Relative to the

earlier discussion

of the figure at

the left:

– Does chemical

looping “fit” this

description?

– What are the

potential benefits

and drawbacks

of chemical

looping w/r to

efficiency?

• What applications

of chemical

looping are most

attractive?

Generator Carbon

Separation

Unit

Q = mfDH Fuel Heat

Input

hg Generator

Efficiency

Wo Gross

Generator

Work

W1 Net

Output

Define:

a = (kg CO2 produced) / (kg fuel burned )

wCO2 = (separation work, Joules ) / (kg CO2) CO2

Approx Ranges: (30 – 60%) (6-10%)

‹#›

Chemical Looping Summary

• Not completely new, but new interest because of CO2 capture.

• Various metal/metal oxide pairs are candidate oxygen carriers.

• For a given heat output, reactor circulation rates depend on the

oxygen capacity and the thermal balance.

• Practical experience from CFB and FCC applications suggest

the range of application.

• In progress:

– kinetic improvements

– reactor design and optimization

– model validation

– attrition

CO2 + H2O

Ash

Recycle

CO2 + H2OFuel

Air

Seal

Seal

N2 + O2

(vitiated air)