fuel cell technology status - voltage degradation · fuel cell technology status . voltage...

NREL is a national laboratory of the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, operated by the Alliance for Sustainable Energy, LLC.

Fuel Cell Technology Status Voltage Degradation

2013 DOE Annual Merit Review

Jennifer Kurtz, Huyen Dinh, Sam Sprik, Genevieve Saur, Chris Ainscough, Mike Peters

May 14, 2013 This presentation does not contain any proprietary, confidential, or otherwise restricted information.

Project ID# FC081

2

Overview

Timeline Project start date: July 2009 Project end date: October 2013* Percent complete: On-going

Barriers Durability of state-of-the-art fuel cell stacks and systems

Budget Total project funding

DOE share: $450k Contractor share: $0

Funding received in FY13: $150k Funding received FY09–12: $300k

Partners 68 fuel cell developers contacted 15 fuel cell developers shared data

*Project continuation and direction determined annually by DOE

3

Relevance: Objectives Benchmark state-of-the-art fuel cell durability

o Develop snapshot of state-of-the-art fuel cell durability o Uniformly apply analysis method to data accumulated in lab o Obtain independent assessment and status of state-of-the-

art fuel cell technology

Leverage analysis experience o Utilize analysis methods, experience, and data from fuel cell

field demonstrations (e.g., DOE’s FCV Learning Demonstration and Early Market demonstrations)

o Compare lab and field data

Collaborate with key fuel cell developers o Provide feedback to fuel cell developers o Investigate factors affecting fuel cell durability o Study differences between lab and field durability o System price benchmarking

All data is supplied voluntarily and published results are updated annually.

4

CDPs

DDPs

Public

Composite Data Products (CDPs) • Aggregated data across multiple systems,

sites, and teams • Publish analysis results once a year without

revealing proprietary data2

Detailed Data Products (DDPs) • Individual data analyses

• Identify individual contribution to CDPs • Shared once a year only with the partner

who supplied the data1

1) Data exchange may happen more frequently based on data, analysis, and collaboration 2) Results published via NREL technology validation website, conferences, and reports (http://www.nrel.gov/hydrogen/proj_learning_demo.html)

Approach: Analysis and Reporting of Real-World Operation Data

Results

Bundled data (operation & maintenance/safety) delivered

to NREL annually Internal analysis

completed annually in HSDC

5

CDP and DDP Schedule

Op Hr Proj Hr Op Hr Proj Hr Op Hr Proj Hr Op Hr Proj Hr0

5000

10000

15000

20000

25000

30000

35000

40000

45000

49% of data setsare short stacks2

Backup


Automotive


Forklift


Prime

Hou

rs

25th-75th PercentileMax Op HrAvg Op HrMax Proj Hr3

Avg Proj Hr3

DOE Target4

NREL cdp_lab_01

Created: Mar-30-12 4:20 PM

(1) At least 9 fuel cell developers supplied data. Analysis will be updated periodically.(2) PEM & SOFC data from lab tested, full active area short stacks and systems with full stacks. Data generated from constant load, transientload, and accelerated testing between 2004 and early 2011.(3) The DOE 10% voltage degradation metric is used for assessing voltage degradation; it may not be the same as end-of-life criteria and doesnot address catastrophic failure modes.(4) DOE targets are for real-world applications; refer to Hydrogen, Fuel Cells, & Infrastructure Technologies Program Plan.

Op Hr Proj Hr

Automotive

Example DDPs

Sample Data Set 1 Sample Data Set 2

0 200 400 600 800 1000 1200 1400 1600 1800 2000220

230

240

250

260

270

280

290

300

Op Hours

Volta

ge

Voltage vs. Operation Hours at 300A: Vehicle19-Stack1

871

900

847

288

259

0

0.2

0.4

0.6

0.8

1Stack Weight Factors

Wei

ght V

alue

Overal

l

DatQty

DatGap

DR1ci

DR2ci

Created: Oct-09-08 3:01 PM

warm-up time=10 min

pwr rate filt=1000 kW/s

amp rate filt=1000 A/s

pts per fit=2500

1 data pt every 1seconds

0 200 400 600 800 1000 1200 1400 1600 1800 2000220

230

240

250

260

270

280

290

300

Op Hours

Volta

ge


871

900

847

288

259

0

0.2

0.4

0.6

0.8


Wei

ght V

alue

Overal

l

DatQty

DatGap

DR1ci

DR2ci


warm-up time=10 min



pts per fit=2500


Example Data Results (if needed) Draft CDPs

Initial Review

(~1 weeks)

Final Draft CDPs

(< 1 week)

Final Review

(< 1 week)

Finalized CDPs

(< 1 week)

Data Process

& Analysis

(~4 weeks excluding data processing and analysis)

6

• Record operation hour for segment

Approach: Raw FC Data Processing Example Data

• Record voltages from polarization fit at set currents

• Segment fuel cell voltage and current data

• Apply polarization fit 1

2

3

4

1

2

3

4

0 200 400 600 800 1000 1200 1400 1600 1800 2000220

230

240

250

260

270

280

290

300

Op Hours

Volta

ge


871

900

847

288

259

0

0.2

0.4

0.6

0.8


Wei

ght V

alue

Overal

l

DatQty

DatGap

DR1ci

DR2ci


warm-up time=10 min



pts per fit=2500


5 6

7

8

9

5

6

7

8

9

Plot polarization fit voltage at a specific current Apply robust segmented linear fit (if trend suggests non-linear degradation trend) Record fit y-intercept (nominal voltage drop)

Record operation hour when fit crosses 10% nominal voltage drop Investigate fit quality

7

Accomplishment: Annual Update Completed FY09 FY10 FY11 FY12 FY13


5000

10000

15000

20000

25000

30000

35000

40000

45000


Backup


Automotive


Forklift


Prime

Hou

rs

Operation Hours and Projected Hours to 10% Voltage Drop1


Avg Proj Hr3

DOE Target4

NREL cdp_lab_01



10% 15% 20% 25%0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

22000

Voltage Degradation Levels2,6

Proj

ecte

d H

ours

3,4,

5

Durability Lab Data Sets1 - Average projected Hour Sensitivity to Voltage Degradation Levels

Average -ForkliftAverage -PrimeAverage -AutomotiveAverage -Backup

NREL cdp_lab_02


(1) PEM & SOFC data from lab tested, full active area short stacks and systems with full stacks. Data generated from constant load, transient load, and accelerated testing between 2004 and 2011.(2) 10% Voltage degradation is a DOE metric for assessing fuel cell performance.(3) Curves generated using the average of each application at various voltage degradation levels.(4) The projection curves display the sensitivity to percentage of voltage degradation, but the projections do not imply that all stacks will (or do) operate at these voltage degradation levels.(5) Projections may be limited by demonstrated operation hours to minimize extrapolations.(6) The voltage degradation levels are not an indication of an OEM's end-of-life criteria and do not address catastrophic stack failures such as membrane failure.

Gen1 Auto (1) Gen2 Auto (1) Lab Auto (2)0

2000

4000

6000

8000

10000

12000

14000

2006 Target

2009 Target

2015 Target

Proj

ecte

d H

ours

to 1

0% V

olta

ge D

rop

Comparison of Fuel Cell Vehicle Field and Lab Durability Projections

Average Projection3

Max Projection3

NREL cdp_lab_03


(1) Gen1 and Gen2 Data from DOE's Learning Demonstration (2005 - 2010)(2) Lab data providers may not be the same as participants in DOE's Learning Demonstration. 49% of data are full active area short stacks.(3) The DOE 10% voltage degradation metric is used for assessing voltage degradation; it may not be the same as end-of-life criteria and doesnot address catastrophic failure modes.

2008 2009 2010 2011 20120

50

100

150

200

250

Ope

ratio

n H

ours

[1,0

00]

0

20

40

60

80

100

Dat

a Se

t Cou

nt

Cumulative Lab Data Operation Hours and Dates

05/2

009

CD

Ps1

12/2

009

CD

Ps1

06/2

010

CD

Ps1

07/2

011

CD

Ps1

03/2

012

CD

Ps1

Prime Op HrsForklift Op HrsAuto Op HrsBackup Op HrsInactive Data SetsActive Data Sets

NREL cdp_lab_04

Created: Mar-30-12 4:02 PM 1) Data set count at publication of a CDP set - where a data set represents a short stack, full stack, or system test data.

MHE Field (1) Lab MHE (2)0

5000

10000

15000

20000

25000

Proj

ecte

d H

ours

to 1

0% V

olta

ge D

rop

Comparison of Fuel Cell Material Handling Equipment Field and Lab Durability Projections

Average Projection3

Max Projection3

NREL cdp_lab_05


(1) MHE Data from DOE's Early Market Demonstration (2009 - )(2) Lab data providers may not be the same as participants in DOE's Field Demonstration. 20% of data are full active area short stacks.(3) The DOE 10% voltage degradation metric is used for assessing voltage degradation; it may not be the same as end-of-life criteria and doesnot address catastrophic failure modes.

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 500

10

20

30

40

50

60

70

80

90

100

Dat

a Se

t1 [%]

Power Capability2 [kW]

Power Capability of Test Data Sets

BackupAutomotiveForkliftPrime

NREL cdp_lab_06


>50

1) A data set represents a short stack, full stack, or system test data.2) Power capability represents the maximum power for a data set but not necessarily the load profile or time at a power level.

0 1000 2000 3000 4000 5000 6000 7000 8000 9000 100000

5

10

15

20

25

Operation Hours

Dat

a Se

ts1 [%

]

Operation Hours and Data Sets Operated Beyond 10% Voltage Degradation

Data Sets < 10% VdegData Sets2 > 10% Vdeg

NREL cdp_lab_07


>10000

1) A data set represents a short stack, full stack, or system test data.2) The DOE 10% voltage degradation metric is used for assessing voltage degradation; it may not be the same as end-of-life criteria and does not address catastrophic failure modes. Some data sets have operated beyond 10% voltage degradation because they are able to satisfy the operating requirements at a higher percentage of voltage degradation or the test is designed to operate until a failure.

All Short Stack Full Stack System Steady Duty Cycle Accelerated0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

22000Projected Hours to 10% Voltage Degradation by Configuration and Test Condition

Proj

ecte

d H

ours

to 1

0% V

olta

ge D

egra

datio

n1,2

Average -BackupAverage -AutomotiveAverage -ForkliftAverage -Prime

NREL cdp_lab_08


Configurations, All Test Conditions Test Conditions3, All Configurations1) The DOE 10% voltage degradation metric is used for assessing voltage degradation; it may not be the same as end-of-life criteria and does not address catastrophic failure modes.2) Not all applications have data sets in each configuration or test condition group.3) Steady - little or no change to load profile Duty Cycle - load profile mimics real-world operating conditions Accelerated - test profile is more aggressive than real-world operating conditions


5000

10000

15000

20000

25000

30000

35000

40000

45000


Backup


Automotive


Forklift


Stationary

Hou

rs

Operation Hours and Projected Hours to 10% Voltage Drop - Lab Data Sets1


Avg Proj Hr3

DOE Target4

NREL cdp_lab_01

Created: Aug-02-11 6:21 PM


10% 15% 20% 25%0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000


Proj

ecte

d H

ours

3,4,

5

Durability Lab Data Sets1 - Average Projected Hour Sensitivity to Voltage Degradation Levels

Automotive Average ProjectionBackup Average ProjectionForklift Average ProjectionStationary Average Projection

NREL cdp_lab_02


(1) PEM & SOFC data from lab tested, full active area short stacks and systems with full stacks. Data generated from constant load, transient load, and accelerated testing between 2004 and early 2011.(2) 10% Voltage degradation is a DOE metric for assessing fuel cell performance.(3) Curves generated using the average of each application at various voltage degradation levels.(4) The projection curves display the sensitivity to percentage of voltage degradation, but the projections do not imply that all stacks will (or do) operate at these voltage degradation levels.(5) Projections may be limited by demonstrated operation hours to minimize extrapolations.(6) The voltage degradation levels are not an indication of an OEM's end-of-life criteria and do not address catastrophic stack failures such as membrane failure.


2000

4000

6000

8000

10000

12000

14000

2006 Target

2009 Target

2015 Target

Proj

ecte

d H

ours

to 1

0% V

olta

ge D

rop


Average Projection3

Max Projection3

NREL cdp_lab_03



2008 2009 2010 20110

50

100

150

200

250

Ope

ratio

n H

ours

[1,0

00]

0

20

40

60

80

100

Dat

a Se

t Cou

nt


05/2

009

CD

Ps1

12/2

009

CD

Ps1

06/2

010

CD

Ps1

07/2

011

CD

Ps1

Prime Op HrsForklift Op HrsAuto Op HrsBackup Op HrsInactive Data SetsActive Data Sets

NREL cdp_lab_04

Created: Aug-02-11 6:44 PM 1) Data set count at publication of a CDP set


5000

10000

15000

20000

25000

Proj

ecte

d H

ours

to 1

0% V

olta

ge D

rop


Average Projection3

Max Projection3

NREL cdp_lab_05




5000

10000

15000

20000

25000

30000

35000

40000

45000

56% Short Stack Data Sets2

Automotive

33% Data Sets Still Operating


Backup



Forklift



Stationary


Hou

rs

Durability Lab Data Sets1 - Hours


Avg Proj Hr3

DOE Target4

NREL cdplab01

Created: Jun-01-10 4:00 PM


10% 15% 20% 25%0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

22000


Proj

ecte

d H

ours

3,4,

5

Durability Lab Data Sets1 - Average Projected Hour Sensitivity to Voltage Degradation Levels

Automotive Average ProjectionBackup Average ProjectionForklift Average ProjectionStationary Average Projection

NREL cdplab02


(1) PEM & SOFC data from lab tested, full active area short stacks and systems with full stacks. Data generated from constant load, transient load, and accelerated testing between 2004 and early 2010.(2) 10% Voltage degradation is a DOE metric for assessing fuel cell performance.(3) Curves generated using the average of each application at various voltage degradation levels.(4) The projection curves display the sensitivity to percentage of voltage degradation, but the projections do not imply that all stacks will (or do) operate at these voltage degradation levels.(5) Projections may be limited by demonstrated operation hours to minimize extrapolations.(6) The voltage degradation levels are not an indication of an OEM's end-of-life criteria and do not address catastrophic stack failures such as membrane failure.


2000

4000

6000

8000

10000

12000

14000

2006 Target

2009 Target

2015 Target

Proj

ecte

d H

ours

to 1

0% V

olta

ge D

rop


Average Projection3

Max Projection3

NREL cdplab03



Automotive Forklift StationaryApplication Category

Tim

e [H

ours

](5)

0

5000

10000

15000

20000

25000

30000

35000

40000

45000Automotive Target6

Stationary Target6

Created: Dec-01-09 4:29 PM

(1) The DOE 10% voltage degradation metric isa general metric for assessing voltagedegradation, which may not be the same as anOEM's end-of-life criteria and does not address"catastrophic" failure modes such as membranefailure. (2) Collected data consists of lab test data fromshort stacks and systems in multiple applications. (3) Collected data was generated with varioustest bench operation such as constant currentand accelerated degradation tests. (4) Operating period for collected data spansfrom 2004 to early 2009. (5) The upper limit of the bar represents themaximum operating time to 10% voltagedegradation. (6) The DOE Automotive and Stationary targetsare applied to real-world applications; refer toHydrogen, Fuel Cells, & InfrastructureTechnologies Program Plan.

PEM Fuel Cell Stack 10% Voltage Degradation Projections1 -Laboratory Durability Test Data2-4

Op Hr Proj Hr Op Hr Proj Hr Op Hr Proj Hr Op Hr Proj Hr Op Hr Proj Hr Op Hr Proj Hr0

5000

10000

15000

20000

25000

30000

35000

40000

45000

Backup Automotive Bus Forklift Portable

N

ot E

noug

h D

ata

for C

DP

Prime

Hou

rs

Operation Hours and Projected Hours to 10% Voltage Drop1


Avg Proj Hr3

DOE Target4

NREL cdp_lab_01


(1) At least 15 U.S. and international fuel cell developers supplied data. Analysis is updated periodically.(2) PEMFC, DMFC & SOFC data from lab tested, full active area short stacks and systems with full stacks. Data generated from constant load,transient load, and accelerated testing between 2004 and early 2012.(3) The DOE 10% voltage degradation metric is used for assessing voltage degradation; it may not be the same as end-of-life criteria and doesnot address catastrophic failure modes.(4) DOE targets are for real-world applications; refer to Hydrogen, Fuel Cells, & Infrastructure Technologies Program Plan.

10% 15% 20% 25%0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

22000


Proj

ecte

d H

ours

3,4,

5

Durability Lab Data Sets1 - Average projected Hour Sensitivity to Voltage Degradation Levels

Average -ForkliftAverage -PrimeAverage -BusAverage -AutomotiveAverage -Backup

NREL cdp_lab_02


(1) PEM & SOFC data from lab tested, full active area short stacks and systems with full stacks. Data generated from constant load, transient load, and accelerated testing between 2004 and 2012.(2) 10% Voltage degradation is a DOE metric for assessing fuel cell performance.(3) Curves generated using the average of each application at various voltage degradation levels.(4) The projection curves display the sensitivity to percentage of voltage degradation, but the projections do not imply that all stacks will (or do) operate at these voltage degradation levels.(5) Projections may be limited by demonstrated operation hours to minimize extrapolations.(6) The voltage degradation levels are not an indication of an OEM's end-of-life criteria and do not address catastrophic stack failures such as membrane failure.


2000

4000

6000

8000

10000

12000

14000

2006 Target

2009 Target

2015 Target

Proj

ecte

d H

ours

to 1

0% V

olta

ge D

rop


Max Projection3

Average Projection3

NREL cdp_lab_03



2008 2009 2010 2011 2012 20130

50

100

150

200

250

300

Ope

ratio

n H

ours

[1,0

00]

0

20

40

60

80

100

Dat

a Se

t Cou

nt


05/2

009

CD

Ps1

12/2

009

CD

Ps1

06/2

010

CD

Ps1

07/2

011

CD

Ps1

03/2

012

CD

Ps1

03/2

013

CD

Ps1

Backup Op HrsAuto Op HrsBus Op HrsForklift Op HrsPrime Op HrsActive Data SetsInactive Data Sets

NREL cdp_lab_04



5000

10000

15000

20000

25000

Proj

ecte

d H

ours

to 1

0% V

olta

ge D

rop


Max Projection3

Average Projection3

NREL cdp_lab_05



0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 500

10

20

30

40

50

60

70

80

90

100

Dat

a Se

t1 [%]



BackupAutomotiveBusForkliftPrime

NREL cdp_lab_06


>50


0 1000 2000 3000 4000 5000 6000 7000 8000 9000 100000

5

10

15

20

25

Operation Hours

Dat

a Se

ts1 [%

]


Data Sets2 > 10% VdegData Sets < 10% Vdeg

NREL cdp_lab_07


>10000



2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

22000Projected Hours to 10% Voltage Degradation by Configuration and Test Condition

Proj

ecte

d H

ours

to 1

0% V

olta

ge D

egra

datio

n1,2

Average -PrimeAverage -ForkliftAverage -AutomotiveAverage -Backup

NREL cdp_lab_08

Created: Apr-01-13 1:42 PM


0 10 20 30 40 50 60 70 80 90 100

Data Sets [%]

Configurations by Data Set

Backup

Automotive

Bus

Forklift

Prime

Full Stack Full System Short Stack Single Cell

NREL cdp_lab_09

Created: Mar-26-13 11:30 AM

2010 2011 2012 20132

4

6

8

10

12

14

16

18Average Projected Hours to 10% Voltage Drop1,2

Year

Proj

ecte

d H

ours

[1,0

00]

BackupAutomotiveForkliftPrimeBus

NREL cdp_lab_10


(1) The DOE 10% voltage degradation metric is used for assessing voltage degradation; it may not be the same as end-of-life criteria and doesnot address catastrophic failure modes.(2) At least 13 fuel cell developers supplied data, including international. Analysis is updated periodically.

2010 2011 2012 20130

2

4

6

8Backup

Proj

ecte

d H

ours

[1,0

00]

2010 2011 2012 20130

5

10

15Automotive

Proj

ecte

d H

ours

[1,0

00]

2010 2011 2012 20130

5

10

15

20

25Forklift

Proj

ecte

d H

ours

[1,0

00]

2010 2011 2012 20130

10

20

30

40

50

Proj

ecte

d H

ours

[1,0

00]

Prime

Avg Proj HrMax Proj Hr25th-75th Percentile

NREL cdp_lab_11


Projected Hours to 10% Voltage Drop1,2


0 10 20 30 40 50 60 70 80 90 100

Data Sets [%]

Fuel by Data Set

Backup

Automotive

Bus

Forklift

Prime

Hydrogen Reformate

NREL cdp_lab_12


0 10 20 30 40 50 60 70 80 90 100

Data Sets [%]

Test Conditions by Data Set

Backup

Automotive

Bus

Forklift

Prime

Accelerated Duty Cycle Steady

NREL cdp_lab_13


0 10 20 30 40 50 60 70 80 90 100

Backup

Automotive

Forklift

Prime

Data Sets [%]

Current Density Point by Data Set1

<=0.2 [A/cm2] <=0.4 [A/cm2] <=0.6 [A/cm2] <=0.8 [A/cm2] <=1.0 [A/cm2] <=1.2 [A/cm2]

NREL cdp_lab_14

Created: Apr-01-13 10:20 AM 1) Current density referenced are the points at which the voltage degradation is analyzed in CDP Lab 01

The current analysis set is the 5th annual update. The number of data sets, participants, and results have increased each year.

8

2008 2009 2010 2011 2012 20130

50

100

150

200

250

300

Ope

ratio

n H

ours

[1,0

00]

0

20

40

60

80

100

Dat

a Se

t Cou

nt


05/2

009

CD

Ps1

12/2

009

CD

Ps1

06/2

010

CD

Ps1

07/2

011

CD

Ps1

03/2

012

CD

Ps1

03/2

013

CD

Ps1

Backup Op HrsAuto Op HrsBus Op HrsForklift Op HrsPrime Op HrsActive Data SetsInactive Data Sets

NREL cdp_lab_04


Accomplishments: Data Set Count and Operation Hours 04/2013 CDPs include 98 data sets analyzed, 78% retired 16 new data sets since 05/2012 14 CDPs (6 new since 05/2012)

9

* DOE Targets Under Review

* *

Accomplishments: Voltage Degradation Results by Application The average projected times to 10% voltage drop are 2,500, 3,600, 6,200, 14,600, and 9,300 for backup power, automotive, bus, forklift, and stationary applications, respectively.

PEMFC & SOFC data from lab tested, full active area short stacks and systems with full stacks. Data generated from constant load, transient load, and accelerated testing and includes 15 U.S. and international fuel cell developers.

10

Accomplishments: Voltage Degradation by Configuration and Test Condition


2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

22000j g g y g

Proj

ecte

d H

ours

to 1

0% V

olta

ge D

egra

datio

n1,2

Average -PrimeAverage -ForkliftAverage -AutomotiveAverage -Backup

NREL cdp_lab_08



Not all applications have data sets in each configuration or test condition group. Steady – little of no change to load profile Duty Cycle – load profile mimics real-world operating conditions Accelerated – test profile is more aggressive than real-world operating conditions

Updated since 05/2012

Grouping by configuration and test condition is important because of influences on degradation projects. Conclusions are still difficult to identify because of limited data and many variations in these groups. Further study and additional data are needed.

11

Accomplishments: Breakdown of Data Set Configuration Used for Voltage Degradation Results

0 10 20 30 40 50 60 70 80 90 100

Data Sets [%]

Backup

Automotive

Bus

Forklift

Prime

Full Stack Full System Short Stack Single Cell

NREL cdp_lab_09

Created: Mar-26-13 11:30 AM

Configuration is connected to test objective, technology generation, setup, and design. Configuration can have a significant impact on the voltage degradation projects and is an area for continued study and data gathering.

12

Accomplishments: Breakdown of Data Set Test Conditions Used for Voltage Degradation Results

0 10 20 30 40 50 60 70 80 90 100

Data Sets [%]

Backup

Automotive

Bus

Forklift

Prime

Accelerated Duty Cycle Steady

NREL cdp_lab_13


Steady operation is a common protocol for all studied applications. Test condition has a significant impact on the voltage degradation and is an area for continued study and data gathering.

13

Accomplishments: Breakdown of Current Densities Used for Voltage Degradation Results

0 10 20 30 40 50 60 70 80 90 100

Backup

Automotive

Forklift

Prime

Data Sets [%]

y y

<=0.2 [A/cm2] <=0.4 [A/cm2] <=0.6 [A/cm2] <=0.8 [A/cm2] <=1.0 [A/cm2] <=1.2 [A/cm2]

NREL cdp_lab_14

C t d A 01 13 10 20 AM

Current density point at which the projected time to 10% voltage degradation is analyzed varied based on available data, test purpose, and application. Future analysis will look to publish voltage degradation at common current density points.

New since 05/2012

14

Accomplishments: Average Projected Hours to 10% Voltage Degradation by Year

2010 2011 2012 20132

4

6

8

10

12

14

16

18Average Projected Hours to 10% Voltage Drop1,2

Year

Proj

ecte

d H

ours

[1,0

00]

BackupAutomotiveForkliftPrimeBus

NREL cdp_lab_10



Most applications have seen an increase in average time to 10% voltage degradation over the last three years. Averages may decrease from year to year based on new data sets added and varied test conditions and configurations.

15

Accomplishments: Project Information Package

System price data for price benchmarking

Pamphlet with participation details and benefits

Website with published results and presentations Example CDP results

for price data

New since 05/2012

Metadata for voltage degradation analysis

16

Collaborations • Working with multiple fuel cell developers

o 15 of 68 fuel cell developers contacted have supplied at least one dataset o Reasons for developers not providing data include concerns over voluntary

proprietary data sharing, availability of data sets with high operation time that are a good fit to include in the analysis, and readily accessible data in the requested format.

o Data contributors are not identified yet because of limited data sets by application category.

• Gathering PEMFC, DMFC, and SOFC datasets • Data sharing is completely voluntary • Participation in DOE durability working group and presentation of status

to fuel cell tech team • Ongoing effort with fuel cell developers to:

o Include new data sets (particularly in the stationary category) o Update datasets already included if applicable o Include new fuel cell developers

17

Future Work

• Publish report on durability analysis method • Add price information to the project collaboration pamphlet • Continue cultivating existing collaborations and developing new

collaborations with fuel cell developers. • Next planned update is 03/2014 • Possibilities that may be included in the next update are:

o Expand results aimed at improving data comparability and statistical confidence

o Voltage degradation at common and multiple current density points and power points

o Price benchmarking o Results by testing protocol o PEM, SOFC, & DMFC o Single cell o Investigate other aging parameters for fuel cell durability (e.g.

start/stops, soak time)

18

Future Work: Price Data

Objective: Report on status and establish expectation of current price data at low volumes.

1) DOE cost targets for 1 – 10 kW residential (http://www1.eere.energy.gov/hydrogenandfuelcells/mypp/pdfs/fuel_cells.pdf) 2) Data estimate source is from ARRA installations in 2012 dollars. 3) Data estimate source is from public information in 2012 dollars for US and international companies and multiple fuel cell types (≤5kW).

2 2 3

19

Summary Relevance: Independent assessment of state-of-the-art fuel cell technology provides one location for fuel cell durability status from leading fuel cell developers with a uniform analysis and reporting method on a variety of proprietary data.

Approach: Leverage capabilities established under other technology validation activities (NRELFAT) and industry collaborations.

Accomplishments: Fifth annual results were updated for 5 applications plus started portable and include new details based on metadata and durability trends over time and international developers. The data are fully integrated into NRELFAT and an online interface provides information on the project, connection for interested collaborators, and all publications. Improved project information material for details and benefits of participation.

Collaborations and Future Work: Continue expanding analyzed data sets, included fuel cell developers, and results and have price benchmarking.

20

Technical Backup

21

Relevance - Benefits of Lab Fuel Cell Durability Analysis

External • One source of durability status

for DOE from the leading fuel cell developers

• Independent technology assessment with uniform analysis on proprietary data

• Highlights technology successes

• Helps adoption of fuel cell technology

• Identifies areas for continued development

• Possible to study differences between field and lab performance

• Benchmarking against CDPs • Analysis method may provide

a new way to study fuel cell durability

• Collaboration with NREL’s technology validation team; dedicated analysis team with experience in multiple fuel cell applications

• Provide information that is very useful for external partners (e.g. DOE) without revealing proprietary information

• Possible to study differences between field and lab performance

Internal

22

Accomplishments: Voltage Degradation by Year

2010 2011 2012 20130

2

4

6

8Backup

Proj

ecte

d H

ours

[1,0

00]

2010 2011 2012 20130

5

10

15Automotive

Proj

ecte

d H

ours

[1,0

00]

2010 2011 2012 20130

5

10

15

20

25Forklift

Proj

ecte

d H

ours

[1,0

00]

2010 2011 2012 20130

10

20

30

40

50

Proj

ecte

d H

ours

[1,0

00]

Prime

Avg Proj HrMax Proj Hr25th-75th Percentile

NREL cdp_lab_11


Projected Hours to 10% Voltage Drop1,2


23

Accomplishments: Data Set Power Capability

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 500

10

20

30

40

50

60

70

80

90

100

Dat

a Se

t1 [%]



BackupAutomotiveBusForkliftPrime

NREL cdp_lab_06


>50


The power capability illustrates the range of fuel cell power for the data sets by application and that many of the data sets are low power either because of short stacks or because high power is not required.

1) A data set represents a short stack, full stack, or system test data. 2) The power capability represents the maximum power for a data set, not the load profile

or time at a power level.

24

Accomplishments: Data Set Operation Hours and the Percentage of Data Sets That Have Passed 10% Voltage Degradation

0 1000 2000 3000 4000 5000 6000 7000 8000 9000 100000

5

10

15

20

25

Operation Hours

Dat

a Se

ts1 [%

]


Data Sets2 > 10% VdegData Sets < 10% Vdeg

NREL cdp_lab_07


>10000


Many data sets have not (or did not) pass the metric of 10% voltage degradation. Data sets operated beyond 10% voltage degradation because end-of-life criteria may be greater than 10% voltage degradation or the test is designed to operate until a failure.

1) A data set represents a short stack, full stack, or system test data. 2) The DOE 10% voltage degradation metric is used for assessing voltage degradation; it

may not be the same as end-of-life criteria and does not address catastrophic failures.

fuel cell technology status - voltage degradation · fuel cell technology status . voltage...

Documents