Estimating Future Costs of Power Plants with CO Capture

Estimating Future Costs of Estimating Future Costs of Power Plants with COPower Plants with CO22 CaptureCapture

Edward S. RubinDepartment of Engineering and Public Policy

Carnegie Mellon UniversityPittsburgh, Pennsylvania

Presentation toFifth Annual Conference on Carbon Capture & Sequestration

Alexandria, VirginiaMay 10, 2006

E.S. Rubin, Carnegie Mellon

AcknowledgementsAcknowledgements• Matt Antes• Sonia Yeh• Mike Berkenpas

• John Davison

• Dale Simbeck• Howard Herzog• Jon Gibbins• Keywan Riahi• Leo Schrattenholzer• Rodney Allem

• IEA GHG (John Davison)

• Co-Authors:

• Advisors:

Sponsor:


Study ObjectiveStudy Objective

• Develop an empirically-based method to estimate future costs of power plants with CO2 capture, suitable for use in large-scale energy modeling, R&D planning, and other related efforts


Two Approaches to Estimating Two Approaches to Estimating Future Technology CostsFuture Technology Costs

• Method 1: Engineering-Economic ModelingA “bottom up” approach based on engineering process models, informed by expert elicitations regarding potential improvements in key process parameters

• Method 2: Use of Historical Experience CurvesA “top down” approach based on use of mathematical “learning curves” or “experience curves” reflecting historical trends for analogous technologies or systems

This study employs the latter method


Study ApproachStudy Approach

• Quantify historical learning rates of energy and environmental technologies relevant to power plants with CO2 capture

• Apply these results to leading plant design options to estimate learning rates and future plant costs

Note: This study does not include the costs of CO2 transport and storage technologies


• Detailed report available from International Energy Agency Greenhouse Gas Programme(IEA GHG)

Retrospective Case StudiesRetrospective Case Studies



Case Study TechnologiesCase Study Technologies

• Flue gas desulfurization systems (FGD)

• Selective catalytic reduction systems (SCR)

• Gas turbine combined cycle system (GTCC)

• Pulverized coal-fired boilers (PC)

• Liquefied natural gas plants (LNG)

• Oxygen production plants (ASU)

• Hydrogen production plants (SMR)


Learning Curve Formulation Learning Curve Formulation

where,yi = time or cost to produce ith unitxi = cumulative production thru period ib = learning rate exponenta = coefficient (constant)

General equation:yi = axi

–b

Percent cost reduction for a doubling of cumulative output is called the “learning rate” (LR) = (1 – 2 –b)


FGD System Capital CostsFGD System Capital Costs

(Based on 90% SO(Based on 90% SO22 removal, 500 MW plant, 3.5%S coal)removal, 500 MW plant, 3.5%S coal)

10%

100%

1 10 100 1000Cumulative World Capacity of Wet FGD Systems (GWe)

FGD

Cap

ital C

ost

(% o

f bas

e va

lue) y = 1.45x -0.168

R 2 = 0.79

19761980

19821990

1995

Cost reduction = 11%per doubling of

installed capacity;50% reductionover 20 years

10%

100%

1 10 100 1000Cumulative World Capacity of Wet FGD Systems (GWe)

FGD

Cap

ital C

ost

(% o

f bas

e va

lue) y = 1.45x -0.168

R 2 = 0.79

19761980

19821990

1995


installed capacity;50% reductionover 20 years


SCR System Capital CostsSCR System Capital Costs

(Based on 80% NO(Based on 80% NOxx removal, 500 MW plant, medium S coal)removal, 500 MW plant, medium S coal)

10%

100%

1 10 100Cumulative World Capacity of SCR at Coal-Fired Plants (GWe)

SCR

Cap

ital c

ost (

% o

f bas

e va

lue)

y = 1.28x-0.18

R 2 = 0.75

1983 1989

1996

19951993


installed capacity

10%

100%

1 10 100Cumulative World Capacity of SCR at Coal-Fired Plants (GWe)

SCR

Cap

ital c

ost (

% o

f bas

e va

lue)

y = 1.28x-0.18

R 2 = 0.75

1983 1989

1996

19951993


installed capacity


Early Trend of FGD Capital CostEarly Trend of FGD Capital Cost

1 9 6 81 9 7 2

1 9 7 4

1 9 7 5

1 9 7 61 9 8 0

1 9 8 2

1 9 9 0

1 9 9 5

0

5 0

1 0 0

1 5 0

2 0 0

2 5 0

3 0 0

0 1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0 1 0 0C u m u lative W or ld W e t FGD In s tal l e d

C apac i ty (GW )

Cap

ital C

osts

($/k

W) i

n 19

97$

( 1 0 0 0 M W , e f f = 8 0 - 9 0 % )

( 2 0 0 M W , e f f = 8 7 % )

Initial cost estimates were a bit optimistic (O&M costs also low)


Early Trend of SCR Cost EstimatesEarly Trend of SCR Cost Estimates

1983

1989

1993

1995 2000

-20 -10 0 10 20 30 40 50 60 70 80

Cumulative World SCR Installed Capacity (GW)

SCR

Cap

ital C

osts

($/k

W),

1997

$ First Japan commercial installation on a coal-fired power plant

← First German commercial installation

↓ First US commercial installation

←

40

50

60

70

80

90

100

110

120

1977

1978

1979

1980

(early O&M costs also low)


GTCC Capital CostsGTCC Capital Costs

PR~75% (1991-1997)

PR>100% (1981-1991)

Cumulative installed capacity, MW

Inve

stm

ent P

rice

USD

(199

0$/k

W)

10000 100000

300

400

500

600

700800900

1000

Gas Turbine Combined Cycles

Source: Colpier and Cornland (2002).


LNG Plant Capital CostsLNG Plant Capital Costs

y = 269x-0.22

R2 = 0.52

10.0

100.0

1000.0

0.1 1 10 100

Cumulative LNG produced (Mta)

Liqu

efac

tion

capi

tal c

ost (

$/tp

a)

Actual liquefaction unit costTheoretical liquefaction unit costLNG Production

y = 269x-0.22

R2 = 0.52

10.0

100.0

1000.0

0.1 1 10 100

Cumulative LNG produced (Mta)

Liqu

efac

tion

capi

tal c

ost (

$/tp

a)

Actual liquefaction unit costTheoretical liquefaction unit costLNG Production


PC Boiler Capital CostsPC Boiler Capital Costs

y = 515.00x-0.08

PR = 0.95

100

1000

0 1 10 100 1000 10000Cumulative World Pulverized-Coal Plant Installed Capacity (GW)

Subc

ritic

al P

C U

nit C

ost (

1994

$/

kW)

1942, EF=29.9%

19651999, US DOEEF=37.6%

Pulverized Coal-Fired Boilers


Oxygen Plant Capital CostOxygen Plant Capital Cost

y = 94254x-0.157

R2 = 0.43

10000

100000

1000 10000 100000

Cumulative Oxygen Production since 1980(Billion cubic feet)

Rea

l cap

ital c

ost

(US

$200

3/tp

d)

Oxygen Production


Hydrogen Plant Cost Trends Hydrogen Plant Cost Trends

y = 2174.30x-0.47

R2 = 0.65

y = 724.77x-0.47

R2 = 0.65

1

10

100

1,000 10,000 100,000 1,000,000Estimated Cumulative World Hydrogen Produced by SMR (billion

cubic feet)

H2 U

nit P

rice

(200

0 ce

nts

per

hund

red

cubi

c fe

et)

Estimated Capital ChargesEstimated non-fuel O&M cost

Hydrogen Production


Case Study Learning RatesCase Study Learning Rates“Best Estimate”Learning Rates

Technology Capital Cost

O&M Cost

Flue gas desulfurization (FGD) 0.11 0.22

Selective catalytic reduction (SCR) 0.12 0.13

Gas turbine combined cycle (GTCC) 0.10 0.06

Pulverized coal (PC) boilers 0.05 0.18

LNG production 0.14 0.12

Oxygen production (ASU) 0.10 0.05

Hydrogen production (SMR) 0.27 0.27

Results are within ranges reported for other energy-related technologies

Application to Power Plants Application to Power Plants with COwith CO22 CaptureCapture



Power Plants with COPower Plants with CO22 CaptureCapture

• PC plant with post-combustion capture(amine system)

• PC plant with oxyfuel combustion

• NGCC plant with post-combustion capture(amine system)

• IGCC plant with pre-combustion capture(WGS + Selexol)


Baseline Plant Designs Baseline Plant Designs (1)(1)

PC Plant

Oxyfuel Plant

Coal

Air

Steam

Sta

ck

Amine/CO2

CO2

Steam Turbine

Generator

Electricity

Air PollutionControls

(SCR, ESP, FGD)

CO2 CaptureSystem

PC Boiler

Amine/CO2 Separation

MostlyN2

CO2Compression

Amine

CO2 tostorage

to atmosphere

Coal

Air

Steam

Sta

ck

Amine/CO2

CO2

Steam Turbine

Generator

Electricity


(SCR, ESP, FGD)

CO2 CaptureSystem

PC Boiler

Amine/CO2 Separation

MostlyN2

CO2Compression

Amine

CO2 tostorage

to atmosphere

CoalDistillation

water

CO2

Air

H2O

O2

SteamTurbine

Generator

Electricity

Steam

PC Boiler CO2compression

Air Separation

Unit

CO2 to recycle

to atmosphere


(ESP, FGD)

CO2CO2CoalDistillation

water

CO2

Air

H2O

O2

SteamTurbine

Generator

Electricity

Steam

PC Boiler CO2compression

Air Separation

Unit

CO2 to recycle

to atmosphere


(ESP, FGD)

CO2CO2 to storage

Stack


Baseline Plant Designs Baseline Plant Designs (2)(2)

Gas TurbineCombined

Cycle SystemH2O St

ack

O2

Air

Coal

Gasifier

Stac

k

Air Separation

Unit

ShiftReactor CO2

Selexol/CO2Separation

Selexol Selexol/CO2

H2CO2

CaptureSystem

CO2compression

CO2 tostorage

Quench System

H2Sulfur

Removal System

to atmosphere

H2O

Electricity

Air

Gas TurbineCombined

Cycle SystemH2O St

ack

O2

Air

Coal

Gasifier

Stac

k

Air Separation

Unit

ShiftReactor CO2

Selexol/CO2Separation

Selexol Selexol/CO2

H2CO2

CaptureSystem

CO2compression

CO2 tostorage

Quench System

H2Sulfur

Removal System

to atmosphere

H2O

Electricity

Air

NaturalGas S

tack

Amine Amine/CO2

CO2 tostorage

Electricity SteamTurbine

Generator

CO2 CaptureSystem

Steam

GasTurbine

Amine/CO2Separation

MostlyN2

CO2Compression

CO2

SteamGenerator

AirCompressor

Combustor

Air

to atmosphere

NaturalGas S

tack

Amine Amine/CO2

CO2 tostorage

Electricity SteamTurbine

Generator

CO2 CaptureSystem

Steam

GasTurbine

Amine/CO2Separation

MostlyN2

CO2Compression

CO2

SteamGenerator

AirCompressor

Combustor

Air

to atmosphereNGCC Plant

IGCC Plant


Step 1Step 1: Disaggregate each plant : Disaggregate each plant into major subinto major sub--sections sections

For example:

• IGCC Plant ComponentsAir separation unitGasifier areaSulfur removal/recovery systemCO2 capture system (WGS+Selexol)CO2 compressionGTCC (power block)Fuel cost


Step 2Step 2: Estimate current plant costs and : Estimate current plant costs and contribution of each subcontribution of each sub--section section

Plant Type & Technology Capital Cost

Annual O&M Cost*

Cost ofElectricity*

IGCC Plant w/ Capture 1,831 $/kW 21.3 $/MWh 62.6 $/MWhAir separation unit 18 % 8 % 14 %Gasifier area 27 % 17 % 24 %Sulfur removal/recovery 6 % 3 % 5 %CO2 capture system* 13 % 7 % 11 %CO2 compression 2% 2 % 2 %GTCC (power block) 34 % 9 % 25 %Fuel cost** -- 54% 19 %

Levelized costs in constant $2002 Levelized costs in constant $2002

*Excludes costs of CO2 transport and storage **Based on Pittsburgh #8 coal @ $1.0/GJ


Baseline costs obtained from IECMBaseline costs obtained from IECM

• A computer model developed for DOE/NETL, benchmarked on recent engineering studies

• Provides preliminary design estimates of performance, emissions and cost for:

PC, NGCC and IGCC plantsConventional AP controlsCCS options (pre- and post-combustion, oxyfuel comb.)

• Free Web Download :www. iecm-online.com


Step 3Step 3: Select learning rate analogues : Select learning rate analogues for each plant componentfor each plant component

Plant Type & Technology FGD SCR GTCC PC

boilerLNG prod

O2prod

IGCC PlantAir separation unit XGasifier area XSulfur removal/recovery X XCO2 capture system X XCO2 compressionGTCC (power block) X


Step 4Step 4: Estimate current capacity : Estimate current capacity of major plant componentsof major plant components

Plant Type &TechnologyCurrent

MWnet

Equiv.

IGCC Plant ComponentsAir separation units 50,000

Gasifier area 10,000

Sulfur removal/recovery 50,000

CO2 capture system 10,000

CO2 compression 10,000

GTCC (power block) 240,000


Step 5Step 5: Set projection period : Set projection period and start of learningand start of learning

Cumulative CCS Capacity (MW)

Plant Type Learning Begins at:1st Plant nth Plant

Learning Projected

to:NGCC Plant 432 3,000 100,000PC Plant 500 5,000 100,000

IGCC Plant 490 7,000 100,000Oxyfuel Plant 500 10,000 100,000


Step 6Step 6: Sensitivity Analysis: Sensitivity Analysis

• Learning starts at either first or nth plant• Range of component learning rates• Projection to 50 GW of worldwide capacity • Lower estimates of current component capacity• Effect of additional non-CCS experience • Higher fuel prices for coal and natural gas • Lower financing costs + higher plant utilization


Detailed Detailed Results Results

are are AvailableAvailable

Learning Rate

Initial Value

Final Value % Change Learning

RateInitial Value

Final Value % Change

Nominal Base Case Assumptions 0.022 916 817 10.8% 0.033 59.1 49.9 15.5%Learning Starts with First Plant 0.014 916 811 11.5% 0.028 59.1 47.0 20.4%Learning up to 50 GW 0.018 916 849 7.3% 0.031 59.1 52.0 12.0%Current Capture Capacity = 0 GW 0.029 916 786 14.2% 0.037 59.1 48.8 17.4%Non-CSS Exp. Multipliers = 2.0 0.030 916 783 14.4% 0.036 59.1 49.0 17.1%Natural Gas Price = $6.0/GJ 0.022 925 826 10.7% 0.033 76.1 64.2 15.7%FCF = 11%, CF = 85% 0.022 918 820 10.7% 0.034 51.6 43.3 16.1%

Learning Rate

Initial Value


RateInitial Value


Nominal Base Case Assumptions 0.021 1,962 1,783 9.1% 0.035 73.4 62.8 14.4%Learning Starts with First Plant 0.013 1,962 1,764 10.1% 0.024 73.4 60.8 17.2%Learning up to 50 GW 0.018 1,962 1,846 5.9% 0.031 73.4 66.0 10.1%Current Capture Capacity = 0 GW 0.026 1,962 1,744 11.1% 0.042 73.4 60.9 17.1%Non-CSS Exp. Multipliers = 2.0 0.029 1,962 1,723 12.2% 0.068 73.4 60.4 17.8%Coal Price = $1.5/GJ 0.021 1,965 1,786 9.1% 0.035 79.6 68.2 14.3%FCF = 11%, CF = 85% 0.021 1,963 1,785 9.1% 0.039 57.2 48.2 15.7%

Learning Rate

Initial Value


RateInitial Value


Nominal Base Case Assumptions 0.050 1,831 1,505 17.8% 0.049 62.6 51.5 17.7%Learning Starts with First Plant 0.029 1,831 1,448 20.9% 0.032 62.6 48.6 22.4%Learning up to 50 GW 0.044 1,831 1,610 12.1% 0.045 62.6 54.9 12.2%Current Gasifier Capacity = 1 GW 0.057 1,831 1,460 20.3% 0.055 62.6 50.2 19.7%Above + H2-GTCC = 0 GW 0.088 1,831 1,285 29.8% 0.078 62.6 45.9 26.6%Non-CSS Exp. Multipliers = 2.0 0.062 1,831 1,432 21.8% 0.054 62.6 49.5 20.8%Coal Price = $1.5/GJ 0.050 1,834 1,507 17.8% 0.048 68.4 56.6 17.3%FCF = 11%, CF = 85% 0.048 1,832 1,516 17.2% 0.047 47.2 39.2 16.9%

Learning Rate

Initial Value


RateInitial Value


Nominal Base Case Assumptions 0.028 2,417 2,201 9.0% 0.030 78.8 71.2 9.6%Learning Starts with First Plant 0.013 2,417 2,160 10.7% 0.017 78.8 68.6 12.9%Learning up to 50 GW 0.023 2,417 2,291 5.2% 0.025 78.8 74.3 5.8%Current Boiler Capacity = 0 0.054 2,417 2,008 16.9% 0.056 78.8 65.1 17.5%Non-CSS Exp. Multipliers = 2.0 0.038 2,417 2,122 12.2% 0.044 78.8 68.8 12.7%Coal Price = $1.5/GJ 0.028 2,421 2,204 9.0% 0.030 84.7 76.4 9.8%FCF = 11%, CF = 85% 0.028 2,418 2,202 9.0% 0.031 58.8 53.0 9.9%

COE ($/MWh)

COE ($/MWh)

COE ($/MWh)

COE ($/MWh)NGCC Sensitivity Case

Capital Cost ($/kW)

PC Sensitivity CaseCapital Cost ($/kW)

IGCC Sensitivity CaseCapital Cost ($/kW)

Oxyfuel Sensitivity CaseCapital Cost ($/kW)


Results for IGCC Capital Cost Results for IGCC Capital Cost (Assume learning begins at first capture plant)(Assume learning begins at first capture plant)

10

100

1,000

10,000

100 1,000 10,000 100,000

CCS Cumulative Capacity (MWnet)

Cap

ital C

ost (

$/kW

)

Total Air separation unit Gasifier area Sulfur removal/recovery CO2 capture (WGS/selexol) CO2 compression GTCC (power block)

Based on nominal case study assumptions


Summary of Learning Rate Results Summary of Learning Rate Results

0

2

4

6

8

10

Pow

er P

lant

Lea

rnin

g R

ate

(%)

NGCC PC IGCC Oxyfuel

CAPITAL COST

0

2

4

6

8

10

Pow

er P

lant

Lea

rnin

g R

ate

(%)


COST OF ELECTRICITY


Summary of COE ResultsSummary of COE Results

30

40

50

60

70

80

90

Cos

t of E

lect

ricity

($20

02/M

Wh)


FINAL COE

0

5

10

15

20

25

30

Perc

ent R

educ

tion

in C

OE


% REDUCTION


Percentage Reduction in Percentage Reduction in Cost of COCost of CO22 CaptureCapture

Capture cost is the difference between plants with and without capture at any point in time. This cost falls more

rapidly than the total cost of plants with capture.


ConclusionsConclusions

• Future reductions in the cost of power plants with CO2 capture will require not only sustained R&D, but also full-scale deployment to foster learning-by-doing

• Results suggests that IGCC plants with CO2 capture have a potential for larger cost reductions compared to combustion-based plants with capture

• The timing and magnitude of future cost reductions are uncertain; policy drivers will play a key role


CaveatsCaveats

• There are many!

• Please see full report for details.

A spreadsheet model accompanies the report to facilitate analyses with other input assumptions

Thank YouThank You


Documents

Estimating Future Costs of Power Plants with CO Capture