High Performance, Durable, Low Cost Membrane Electrode ......th, 2016 Project ID: FC104 This presentation does not contain any proprietary, confidential, or otherwise restricted information

High Performance, Durable, Low Cost Membrane Electrode Assemblies for

Transportation Applications

Andrew Steinbach3M CompanyJune 8th, 2016

Project ID: FC104

This presentation does not contain any proprietary, confidential, or otherwise restricted information

Project OverviewTimeline

• Project start: 9/1/12• Project end: 8/30/16*

* Project extended from 8/30/15 (no-cost).

BarriersA. MEA DurabilityB. Stack Material & Mfg CostC. MEA Performance

Budget• Total DOE Project Value: $4.606MM*

• Total Funding Spent: $4.346MM*

• Cost Share Percentage: 20%* Includes DOE, contractor cost-share, and FFRDC funds, as of 2/28/16.

Partners• Johns Hopkins Univ. (J. Erlebacher)• Oak Ridge Nat’l Lab. (D. Cullen)• Lawrence Berkeley Nat’l Lab.(A. Weber)• Michigan Technological Univ. (J. Allen)• Freudenberg FCCT (V. Banhardt) • Argonne Nat’l Lab. (R. Ahluwalia)• Los Alamos Nat’l Lab. (R. Mukundan,

R. Borup)• General Motors (B. Lakshmanan)

2

Objective and RelevanceOverall Project Objective: Development of a durable, low-cost, robust, and high performance membrane electrode assembly (MEA) for transportation applications, able to meet or exceed the DOE 2020 MEA targets.

Primary Objectives and Approaches This Year

Barriers Addressed

1. Produce Project Best of Class Components and CCMs (to be used for stack testing) via Continuous, Pilot Mfg. Processes

B. CostC. Performance

2. Validate Performance and Operational Robustness of Project Best of Class MEAs in Short Stack.

B. CostC. Performance

3. Evaluate Project Best of Class MEAs for Performance/Cost Modeling and Durability Under ASTs and Load Cycling.

A. DurabilityB. CostC. Performance

MEA, Catalyst Targets Addressed2020 Target Target Values Obj.

Q/∆T 1.45kW / °C 3,4Cost $7 / kW 3,4

Durability with cycling

5000 hours w/ < 10% V loss 2,3,4

Performance @ 0.8V 300mA/cm2 3,4

Performance @ rated power

1000 mW/cm23,4

PGM Content (both electrodes)

0.125g/kWRATED0.125mgPGM/cm2 3,4

3

Approach, Milestones, and Status v. TargetsApproach: Optimize integration of advanced anode and cathode catalysts, based on 3M’s nanostructured thin film (NSTF) catalyst technology platform, with next generation PFSA PEMs, gas diffusion media, cathode interfacial layers, and flow fields for best overall MEA performance, durability, robustness, and cost.1. Place appropriate emphasis on key commercialization and DOE barriers.2. Through advanced diagnostics, identify mechanisms of unanticipated component interactions

resulting from integration of low surface area, low PGM, high specific activity electrodes into MEAs.

MS ID

QTR

Project MilestoneMS 1.2, 2.2, 4.2, and 5.2 based on Achievement of Multiple

Project Goals (See Backup Slides)

% Complete(Mar. ’16)

BUDGET PERIOD 2 (June ‘14-Aug. ‘16)1.2 15 Comp. Cand. Meet Project Perf./Cost Goals. 98%2.2 15 Comp. Cand. Meet Project Cold-Start Goals. 50% (2 of 4)5.2 15 Comp. Cand. Meet Project Durability Goals. 91% (10 of 11)

4.2 15 Best of Class MEA Meets All Perf./Cost, Cold-Start, and Durability Project Goals 84%

3.2 12 Validation of Integrated GDL/MEA Model With ≥ 2 3M MEAs (Different Anode GDLs). 100%

6.3 15 BOC MEA: Short Stack Eval. Complete. 40%

0 15 Final Short Stack to DOE-Approved Location. 40%

Status Against DOE 2020 Targets

Characteristic 2020 Targets

Status, ’15 / ’16

Q/∆T (kW / °C) 1.45(@ 8kW/g)

1.45(@ 6.5/6.8* kW/g)

Cost ($ / kW) 75 / 5*

(PGM only @ $35/gPt; 0.692V)

Durability with cycling (hours) 5000 In progress

Performance @ 0.8V (mA/cm2) 300 304 / 310*

Performance @ rated power (mW/cm2)

1000 855 / 891*

(0.692V, 1.45kW/°C)

PGM total content (g/kW (rated))

0.125 0.155 / 0.147*

(0.692V, 1.45kW/°C)

PGM total loading (mg PGM / cm2

electrode area)0.125 0.133 / 0.131*

*: 2016 values from 2015(Sept.) Best of Class MEA, which include a cathode interlayer with 16µg-Pt/cm2

4

Accomplishments and Progress

Best of Class Component Integration (Task 4.1):Pilot Scale Component Fabrication

5

• Last year, reported improved low T performance w/ reduced hydrophobic backing treatment.

• ~30m of “X3” produced.

• “X3” anode GDL yielded modest improvement over “X2” and > 2x gain over baseline GDL.

“X3” Anode GDLBOC CCM0.0

0.2

0.4

0.6

0.8

0.0

0.2

0.4

0.6

0.8

1.0

0.0

0.2

0.4

0.6

0.8

1.0

Hydrophobic Treatment Level

No MPL, FE3410

No MPL, FE5082 - Aug15

J (

A/c

m2)

Hydrophobic Treatment Level

34oC Cell Temp.

No MPL

34oC Cell Temp.

w/ MPLDownselect

FE3410 (2014)

FE5082(Spring'15)

FE5082(Aug'15)

2979 Baseline

J (

A/c

m2)

40oC Cell Temp.

w/ MPL

FE3410 (2014)

FE5082(Spring'15)

FE5082(Aug'15)

2979 Baseline

Downselect

w/ 3M MPLJ (

A/c

m2)

Batch40C, 100% RH

1/1.5atmA H2/Air

0.40V

Baseline

2979

X3 X2

“B” Cathode Interlayer

0 10 20 30-0.2

0.0

0.2

0.4

0.6

0.8

SEF Est. IL PGM (g/cm2)

No IL

Lab

Pilot "1"

Pilot "2"

FinalMin

. C

ell V

Du

rin

g

Lo

ad

Tra

nsie

nt

(Vo

lts)

60oC, 100% RH, 1A/cm

2

• Based on performance /durability assessment, downselected type “B” IL.

• ~50m of interlayer produced.

• Load transient performance largely in line with other pilot scale runs.

• Loading analysis: 16±3 µgPGM/cm2

• Pilot CCM improved vs. lab.

• ~30m of BOC CCM produced (two lots).

• Total CCM PGM loading (1st lot): 0.105mgPt/cm2

Lab (5) Pilot (12)Mass Activity (A/mgPGM) 0.28 ± 0.03 0.33 ± 0.03

Specific Area (m2/gPGM) 11.8 ± 1.4 14.5 ± 0.7

0.00 0.25 0.50 0.75 1.00 1.25 1.500.4

0.5

0.6

0.7

0.8

0.9

1.0 Lab (5)

Pilot (12)

Cell

Vo

ltag

e (

Vo

lts)

J (A/cm2)

80/68/68oC,

1.5/1.5atmA

CS2/2.5 H2/Air

QS FF; Ca. Cond. Only


Best of Class Component Integration (Task 4.1): 2015(Sept.) 3M NSTF Best of Class MEA Performance, Robustness

6

0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00

0.6

0.7

0.8

0.9

2015(Sept.) BOC NSTF MEA

Pilot Scale CCM ("A")

0.131mgPt

/cm2

6.8kW/g

J (A/cm2)

90oC Cell T., 7.35/7.35psig H

2/Air (OUT), CS(2,100)/CS(2.5, 167),

GDS(120s/pt, High->low J Only). 2012 BOC: 84/84C A/C Dewpoint

2015 BOC: 84/84C Dewpt for J > 0.4. 68/68C Dewpt for J < 0.4

Best of Class NSTF MEA

ANODE: PtCoMn/NSTF. CATHODE: Pt3Ni

7/NSTF (DEALLOY) + Pt/C Interlayer.

PEM: 3M-S 14 725EW. Anode GDL: X2. Cathode GDL: 2979. FF2

2015(Mar.) BOC NSTF MEA

Lab Scale CCM

0.133mgPt

/cm2

6.5kW/gCe

ll V

olta

ge

(V

olts)

Each curve is average of 2 MEAs

0 20 40 60 80 100-0.2

0.0

0.2

0.4

0.6

0.82015 (Sept.)

BOC MEA

Robustness

Targets

2015 (Mar.)

BOC MEA

Min

imu

m C

ell V

Du

rin

g

Lo

ad

Tra

ns

ien

t (V

olt

s)

Cell Temperature (oC)

2013 Baseline

MEA

150/150kPaA CS2/2 H2/Air. 1A/cm

2

60-80oC: 100%RH 30-40

oC: 0% RH

1.5atmA H2/Air Performance Operational Robustness

• 2015(Sept.) BOC has improved power and kinetic performance than 2015(Mar.) BOC MEA

• Improved pilot CCM

• Sept. BOC MEA yields 0.1-0.2V gain in load

transient robustness over Mar. BOC MEA.• Improved “X3” anode GDL, type “B” cathode IL

MEA Anode Cat PEM Cathode Catalyst CCM Anode GDL/Cathode IL2015(Mar.) PtCoMn/NSTF, 15µg/cm2 3M-S 725EW

14µ w/ add.

Dealloyed PtNi/NSTF, 0.103mg/cm2 Lab "X2"/ 30% "A"(15µg/cm2)

2015(Sept.) PtCoMn/NSTF, 19µg/cm2 Dealloyed PtNi/NSTF, 0.096mg/cm2 Pilot "X3"/ 30% "B"(16µg/cm2)

MEA

Spec. Power (kW/g)

Rated Power

(W/cm2)

1/4 Power

(A/cm2)

Mar. 6.5 0.855 0.304

Sept. 6.8 0.891 0.310


Best of Class Integration Diagnostics (Task 4.2): 2015(Sept.) BOC MEA Differential Cell Evaluation for ANL Model

7

0.0

0.2

0.4

0.6

0.8

1.0

0 1 2 3 40.0

0.2

0.4

0.6

0.8

1.00.0

0.2

0.4

0.6

0.8

1.0

0 1 2 3 40.0

0.2

0.4

0.6

0.8

1.0

6 10

100

212

Base Conditions: 80oC, 100/100% RH, 1.5/1.5atmA H

2/Air (inlet), 1/3SLM

5cm2 Active Area in 50cm

2 cell. Potential control, 5min/pt.

Ce

ll V

olt

ag

e (

Vo

lts

)

% O2 in N

2

1

Cell Temp. (oC)

70-90

6045

3.0

2.5

2.0

1.5

1.0

A, C Pressure (atmA)

A, C %RH

100, 50

75

15030

J (A/cm2)

0 1 2 3 40.0

0.2

0.4

0.6

0.8

1.0

Differential

5cm2 MEA

J (A/cm2)

Non-Differential

50cm2 MEA

1.5/1.5atmA H2/Air

5cm2: 80/80/80

oC, 1/3SLM

50cm2: 90/84/84

oC, CS2/2.5

• High J’s achieved, approaching 3A/cm2 under 1.5atmA H2/Air.• Little evidence of losses due to Ni2+ in PEM.• Good RH and T sensitivity; modest loss at dry or low T operation.

pO2, P, RH, Cell T Sensitivity Studies Differential v. 50cm2 Comparison

Durability over Test

• Excellent agreement suggests differential cell data is representative.

• ~20mV loss at 0.3A/cm2 after ~800 additional testing hours.

• 25% ECSA loss.

• Minor loss of high J performance.

0 1 2 3 40.0

0.2

0.4

0.6

0.8

1.0

J (A/cm2)

80/80/80oC

1.5/1.5atmA H2/Air

1/3SLM,5min/pt

0 1k 2k 3k 4k 5k

Support Cycle 1-1.5V, 500mV/s

60oC, 100% RH, 1.5atmA H2/Air

No IL

"B" IL

"B" IL

No IL

80/68/68C, 1.5/1.5atmA, CS2/2.5, 1.46A/cm2

"B" IL

No IL

Accomplishments and ProgressCathode Interlayer Durability (Task 5):AST Durability Evaluations of “B” Interlayer

Electrocatalyst AST Electrocatalyst AST• ‘15 AMR: “A” IL durability

insufficient – load trans. failed after 10k cycles.

• “B” improved, load trans. fails after 20k cycles.

• H2/Air perf. improvesSupport AST

• “B” passes support AST • No net change in load

transient after 5k cycles• H2/Air performance

increases 80mV at 1.46A/cm2 w/ or w/o IL.

• “A” or “B” ILs w/ baseline NSTF CCM pass DOE durability (activity, H2/Air) targets with either AST.

BOL 10k 20k 30k-0.20.00.20.40.6

0.50

0.55

0.60

0.650.0

0.1

0.2

0.3

No IL

Electrocatalyst Cycle 0.6-1.0V, 50mV/s

"B" IL

Load

Tra

ns. C

ell

V (V

olts

)

"A" IL

60oC, 100% RH, 1.5atmA H2/Air

No IL

80/68/68C, 1.5/1.5atmA, CS2/2.5, 1.46A/cm2

"B" IL

H 2/Air

Cell

V (V

olts

) "A" IL

No IL

"B" IL

Mas

s Ac

tivity

(A/m

g)

"A" IL

Support AST

Baseline CCM: 0.05/0.15PtCoMn, 3M 825EW 20u. X2 Anode GDL

1.5atmA H2/Air1.46A/cm2

1.5atmA H2/Air1.46A/cm2

8

Accomplishments and ProgressMEA Rated Power Durability (Task 5): Load/RH Cycle Evaluation of 2015 (Sept.) BOC MEAs; ANL Durability Modeling of NSTF MEAs

Performance during 3M Load/RH Cycle PtCoMn/NSTF MEA Durability Model (ANL)

• BOC MEAs have completed > 1200 hours of 3M load/RH cycle testing.

• Significant performance decay during cycling due to both reversible and irreversible factors.

• Steady OCV – no PEM breach.• Tests ongoing and diagnostics conducted

periodically.

0 1000 20000.0

0.2

0.4

0.6

0.8

1.0

0.8A/cm2

64/68oC,CS1.7/1.9

1.5/1.5atmA

0.2A/cm2

64/64oC,CS5/5

1.5/1.5atmA

Time (hours)

MEA1 MEA2

Cell

V (V

olts

)

OCV85/85oC

1.7/1.5atmA

Station issue

80°C Cell

• NSTF rated power degradation correlates to PFSA decomposition extent (’15 AMR). • 800 hours to predicted 10% irreversible voltage loss; 2300

hours to 20% loss.• Critical requirement for 5000 hour durability (10% V loss):

limit cathode F- to 0.7µgF-/cm2 (80% FER reduction)• Path: Decrease MEA FER and increase cathode durable

activity and surface area.

0.4

0.5

0.6

0.7

0.8

0.9

0.0 0.5 1.0

Cel

l Vol

tage

, VCurrent Density, A.cm-2

Initial200 h400 h800 h

2200 h

High Durability Operation

1400 h

3000 hBaseline MEA0.15PtCoMn/NSTF cathode825EW 20µ PEM

FC17

9


10

Best of Class Component Integration (Task 4.1):New, Improved Durability NSTF “M” Cathode Integration

0.0 0.5 1.0 1.50.5

0.6

0.7

0.8

0.9

1.0Stabilzed NPTF, 0.091mg

PGM/cm

2 BOL

15k

30k

Ce

ll V

olt

ag

e (

Vo

lts

)

J (A/cm2)

80/68/68oC

1.5/1.5atmA H2/Air

CS2/2.5, 2min/pt

Durability (Baseline Format)

• First durable “M” catalyst meets

DOE electrocatalyst AST

durability targets.

• BOC cathode: ~65% mass

activity loss.

Performance (BOC Format)

• “M” integrated into BOC MEA format with expected performance and

operational robustness.

• As compared project BOC MEA, first “M” BOC MEA has:• Lower BOL mass activity (0.28 v. 0.39A/mg)

• Lower rated and specific power (5.7 v. 6.8 kW/g)

• Similar/improved operational robustness

• Additional durable catalysts w/ improved activity from project FC143

will be evaluated in BOC format as available.

Robustness (BOC Format)

0.0 0.5 1.0 1.5 2.00.5

0.6

0.7

0.8

0.9

1.0

NPTF "M" Cathode

0.13mgPGM

/cm2

90oC Cell, GDS(2min/pt)

1.5/1.5atmA (OUT) H2/Air, CS2/2.5

J>0.4: 84/84C DP. J < 0.4: 68/68C DP

Ce

ll V

(V

olt

s)

J (A/cm2)

2015(Sept.) BOC MEA

0.131mgPGM

/cm2

MEA

0 20 40 60 80 100-0.2

0.0

0.2

0.4

0.6

0.8"M"

2015 (Sept.)

BOC MEA

Robustness

Targets

2015 (Mar.)

BOC MEA

Min

imu

m C

ell V

Du

rin

g

Lo

ad

Tra

ns

ien

t (V

olt

s)


2013 Baseline

MEA

150/150kPaA CS2/2 H2/Air. 1A/cm

2

60-80oC: 100%RH 30-40

oC: 0% RH

Both include “X3” anode GDL and “B” cathode

interlayer (16µg/cm2 Pt/C). Loadings are total MEA.

Metric Change/DOE TgtMass Activity (A/mg) -40±0.8 / -40%

V @ 0.8A/cm2 -28±1.4 /-30mVSurf. Area (m2/g) -14±0.15 / NA %

Cathode:

0.091mgPGM/cm2

Proj. FC143, “M”

Accomplishments and ProgressBOC MEA Short Stack Evaluation (Task 6): Single Cell Testing at 3M, GM – H2/Air Performance

At 1.5 A/cm2, order of performance 3M Baseline > 3M March BOC

BOC mass activity, specific areas largely in-line w/ 3M expectation.

3M baseline MEA data consistent w/ GM BOC MEA data inconsistent at high current

(GM: 60mV lower @ 1.5A/cm2)

GM 80°C H2/Air Performance 3M 90°C H2/Air Performance

3M Baseline MEA: 0.05/0.15PtCoMn, 3M 825EW 20µ. 3M 2979/2979 GDL3M BOC MEA: 0.019/0.096PtNi, 3M 725EW 14µ. 3M “X3”/2979+”B”IL GDL

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6

HFR

(Ohm

cm

2 )

Cel

l Vol

tage

(V)

Current Density (A/cm2)

3M-HCT Polarization CurvesH2/Air, 80°C, 60% RHinlet, 150 kPaaoutlet, Stoic (2/2.5)

GM Baseline3M Baseline3M Best-of-Class

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.60.5

0.6

0.7

0.8

0.9

1.0

3M Baseline MEA0.20mgPGM/cm2

(2979/2979)

HFR

(ohm

-cm

2 )

Cell

Volta

ge (V

olts

)

J (A/cm2)

90oC, 1.5/1.5atmA H2/Air, CS(2,100)/CS(2.5, 167)2min/pt.Upscan (high->low J) only. OUTLET P CONTROL84/84oC DP for J > 0.4A/cm2; 68/68oC DP for J < 0.4A/cm2;

3M Sept. BOC MEA0.131mgPGM/cm2

(X2/2979+B)

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

11

Accomplishments and ProgressBOC MEA Short Stack Evaluation (Task 6): Single Cell Testing at 3M, GM – Load Transients

Load transient results for 3M baseline and BOC MEAs similar between sites. Robustness improvement of BOC MEA largely confirmed at single cell level

GM

Mea

sure

men

ts

-0.2

0.0

0.2

0.4

0.6

0.8

0 20 40 60 80 100

Inst

anta

neou

s C

ell V

olta

ge (V

)

Cell Temperature (°C)

0% RH

GM Baseline3M BOC

3M Baseline

0 20 40 60 80 100-0.2

0.0

0.2

0.4

0.6

0.8

140/140% RH, 150/150kPa H2/Air, 696/1657SCCM.

V @ 1A/cm2, immediately after J step

3M Baseline MEAs

3M BOC MEAs

Inst

anta

neou

s Ce

ll Vo

ltage

(Vol

ts)


-0.2

0.0

0.2

0.4

0.6

0.8

0 20 40 60 80 100In

stan

tane

ous

Cel

l Vol

tage

(V)


100% RHGM Baseline

3M BOC

3M Baseline-0.2

0.0

0.2

0.4

0.6

0.8

0 20 40 60 80 100

Inst

anta

neou

s C

ell V

olta

ge (V

)


140% RHGM Baseline

3M BOC

3M Baseline

0 20 40 60 80 100

3M Baseline MEAs

3M BOC MEAs


0/0% RH, 150/150kPa H2/Air, 696/1657SCCM.


0 20 40 60 80 100

3M Baseline MEAs

3M BOC MEAs


100/100% RH, 150/150kPa H2/Air, 696/1657SCCM.


3M M

easu

rem

ents

140% RH, 60-80°C 100% RH, 60-80°C 0% RH, 30-50°C

12

Accomplishments and ProgressBOC MEA Short Stack Evaluation (Task 6): Stack H2/Air Performance Much Lower than Expected

Cell voltage of 3M CCMs from short stack testing have been normalized to the GM Baseline performance

3M BOC CCMs show a ~3% improvement in low current density performance over the 3M Baseline Cells

At 1.2 A/cm2, performance of 3M BOC CCMs are ~70% of that observed with GM Baseline Cells

100

150

200

250

300

350

400

40

50

60

70

80

90

100

0 0.2 0.4 0.6 0.8 1 1.2 1.4

Percent HFR

Norm

alized to GM

Baseline

Perc

ent C

ell V

olta

ge n

orm

aliz

ed to

GM

Bas

elin

e

Current Density (A/cm2)

3M-HCT Polarization CurveH2/Air, 80°C, 60% RHinlet, 150 kPaaoutlet, Stoic (2, 2.5)

GM Baseline (X2/2979)3M Baseline (X2/2979) - Normalized3M March BOC (X2/2979 + 'B') - Normalized3M Baseline w/14µm Membrane (X2/2979) - Normalized

Short Stack – 3M-HCT Polarization CurveH2/Air, 80°C, 60% RHinlet, 150 kPaainlet, Stoic (2/2.5)

HFR measuredat 5 kHz frequency

13

Accomplishments and ProgressBOC MEA Short Stack Evaluation (Task 6): Stack Load Transient Performance Much Lower than Expected

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28

Cel

l Vol

tage

V1s

(V)

Cell Number

Instantaneous Response (~1 second)

GM Baseline

NSTF Baseline

GM Baseline

NSTF BOC

NSTF Baselinew/ 14µm Membrane

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28C

ell V

olta

ge V

30s

(V)

Cell Number

Voltage Response at 30 secondsGM Baseline

NSTF Baseline

GM Baseline

NSTF BOC

NSTF Baselinew/ 14µm Membrane

100% RHinlet, 70°C, H2/Air, 150 kPaainlet, Transient from 0.02 to 1.0 A/cm2

At 70°C/100% RH, BOC MEA instantaneous voltage after load transient was negative (failed) Most BOC cells did not recover to positive cell voltages within 30 seconds. Unexpectedly, NSTF baseline CCMs (no cathode interlayer) pass

More optimization of NSTF MEAs with the stack platform is needed to be competitive with conventional catalyst for steady state

and load transient performance.14

Accomplishments and ProgressBOC MEA Short Stack Evaluation (Task 6): Possible Cause for Low Stack Performance: Stack Conditioning Method May Not Be Effective

For a given stack current density, NSTF BOC Cells show lower cell voltages than the Stack Average Voltage

This scenario did not change during the entire stack test duration

Not much improvement in performance observed after reverse flow conditioning

Ineffective or a non-optimized thermal cycle break-in process at the stack level, likely prevents the effective conditioning and performance of the NSTF BOC Cells

0.2

0.4

0.6

0.8

0 25 50 75 100 125 150 175

Volta

ge (V

)

Cycle Count

Stack Average VoltageGM BaselineNSTF BOCNSTF Baseline

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.40 25 50 75 100 125 150 175

Cur

rent

Den

sity

(A/c

m2 )

Cycle Count

Normal Flows ReverseFlows Normal Flows

H2/Air, 75°C, 81% RHinlet, 100 kPaainlet

15


Best of Class Integration Diagnostics (Task 4.2): Best of Class MEA Activation Required for Entitlement Performance, Robustness

16

• Non-optimized “thermal cycle” protocol used to activate BOC MEAs.• Slow voltage and temperature cycles (1.5 hours per complete cycle).• Both electrodes are activated (operated as cell cathodes)

• After anode activation, substantial improvement in 80C H2/air performance and at low temperature.

• ~60mV at 1.5A/cm2, 2x J between 35-50C

Cathode Activation

0 10 20 30 400.0

0.5

1.0

1.5

2.0

75/70/70oC, 1/1atmA H

2/Air

800/1800SCCM, PDS(10s/step)

2015(Sept.) BOC

Time (hours)

J @

0.3

0V

(A

/cm

2) 2015(Mar.) BOC

0.10PtNi/NSTF

as Cathode

50 60 70 Time (hours)

0.02PtCoMn/NSTF

as Cathode

80 90

0.10PtNi/NSTF

as Cathode

Time (hours)

H2/Air Performance

Anode Act. Act. Complete

Temp. Sensitivity

Impact of Anode Act (Sept.)

30 35 40 45 500.0

0.4

0.8

1.2

1.6After Anode

Activation

Before Anode

ActivationJ @

0.4

V (

A/c

m2)

Cell T (oC)

xoC, 100/100% RH,

1.0/1.5atmA H2/Air,

800/1800SCCM,

PSS(0.40V, 5min);

final 1min avg.

0.0 0.5 1.0 1.5 2.00.5

0.6

0.7

0.8

0.9

1.0

Before Anode

Activation

1.5/1.5atmA H2/Air, CS2/2.5

Before: 80/68/68C

After: 90/84/84C

Ce

ll V

(V

olt

s)

J (A/cm2)

After Anode

Activation

Likely contributors to poor stack performance to date:1. “Thermal cycle” protocol not effective in GM stack.2. Slower activation of Sept. BOC MEAs (cause under investigation)


Best of Class Integration Diagnostics (Task 4.2): Rapid NSTF MEA Activation Development

17

• BOC MEA anode HOR kinetics are highly deactivated prior to conditioning.• Challenge w/ very low SEF

anode (~2.5cm2Pt/cm2

planar)• Actual BOC anode deactivation

much larger than predicted.

• Rapid protocol has potential for activation in < 1 hour; simple.

Thermal Cycle (15hrs) Rapid Activation (1hr)

0.0 0.5 1.0 1.5 2.0-0.10

-0.05

0.00

After Anode

Conditioning

Before Anode

Conditioning

J (A/cm2)

80oC Cell, 100% RH,

1.5atmA , 1750SCCM H2

GDS(2min/pt).

IR-F

ree C

ell V

(V

olt

s)

0.0 0.5 1.0 1.5 2.0-0.10

-0.05

0.00

After Rapid

Anode

Conditioning

Before Anode

Conditioning80

oC Cell, 100% RH,

1.5atmA , 1750SCCM H2

GDS(2min/pt).

J (A/cm2)

IR-F

ree C

ell V

(V

olt

s)

0.0 0.5 1.0 1.5 2.00.4

0.6

0.8

1.0

J (A/cm2)

Unactivated

Therm. Cyc. - 5 hrs)

Therm. Cyc. - 6.5hrs

Therm. Cyc. - 40hrs

Ce

ll V

(V

olt

s)

80oC, 1.5atmA H

2/Air

Thermal Cycle (40hrs)

0.0 0.5 1.0 1.5 2.00.4

0.6

0.8

1.0

80oC, 1.5atmA H

2/Air

J (A/cm2)

Therm. Cyc. - 40 hrs

Unactivated

Fast Act. - 1 hrs

Fast Act. - 4 hrs

Fast Act. - 7 hrs

Ce

ll V

(V

olt

s)

Rapid Activation (4hrs) • Dealloyed PtNi/NSTF (not BOC) can be substantially activated in ca. 4 hours.• Optimized T and V cycles• No water injection, no

shutdowns (system friendly).

• To date, has not achieved entitlement performance.

• Method will be optimized for BOC MEA.

Cathode activation work under “FC Fundamentals at Low and Subzero Temp” (Weber)

Rapid, stack friendly activation development is key remaining integration task

An

od

e A

ctiv

atio

nC

ath

od

e A

ctiv

atio

n

Response To Reviewers’ Comments• “Very good progress and a good deal of work have been accomplished in the last year.

Key procedures to improve … apparently promise that, in subsequent research efforts, mostof the performances will be at 2020 levels. However, several problems … are notcompletely solved: the necessity of using an interlayer with the thickness of anothercatalytic layer, the dealloying procedure, and the problem with Ni leaching due to theKirkendall effect. The third point has not been considered at all.”• Agreed, additional materials (e.g. interlayer) and additional processing (e.g. dealloying) are undesirable

from a cost/yield perspective. Tradeoff analysis for this approach vs. others, considering power density,durability, cost (material, process, yield), and end-use system requirements, is complex and not readilyassessable.

• Interlayer thickness is ca. 1-2 µm thick (low loading) and imposes little apparent transport loss(compare 2015(Jan.) to 2015 (Mar.) BOC in ‘15 AMR presentation for interlayer transport impact).

• Transition metal stability in porous metal electrocatalyst has always been a key concern, but believed to betractable. “M” is promising (see FC143).

• “… the key issues have still not been mitigated sufficiently to enable this new catalyst toreplace more conventional MEAs, primarily because 3M has not been willing to make anysignificant changes to the original NSTF catalyst layer structure. …”• Our assessment remains that the project approach will achieve sufficient robustness in stack format.• Per the 2011 FOA, component development was not allowed; our work here was limited to modest

modifications of existing components using known means. Development of a modified NSTF electrodestructure was deemed out of scope.

• An “advanced” NSTF electrode with improved intrinsic robustness is in early development (outside thisproject). Many factors not yet understood, but operational robustness is improved.

18

Collaborations3M – Project management; Materials and process optimization; MEA integration• A. Steinbach, D. van der Vliet, C. Duru, D. Miller, I. Davy, M. Kuznia (Core)

• Cathode Integration: A. Hester, D. Lentz, S. Luopa, D. Tarnowski, B. Smithson, C. Studiner IV, A. Armstrong, M. Stephens, J. Bender, M. Brostrom

• PEM Integration: M. Yandrasits, D. Peppin, G. Haugen, R. Rossiter• Anode GDL/Cathode IL: M. Pejsa, A. Haug, J. Abulu, J. Sieracki• Durability: A. Komlev

Michigan Technological University – GDL char. and PNM modeling; model integration• J. Allen, E. Medici, V. Konduru, C. DeGrootJohns Hopkins University - Pt3Ni7/NSTF dealloying method studies• J. ErlebacherLawrence Berkeley National Laboratory – GDL char. and MEA modeling; model integration• A. Weber, J. MacDonald, I. Zenyuk, A. Kusoglu, S. ShiOak Ridge National Laboratory – Materials characterization (TEM, XPS)• D. Cullen, H. Meyer IIILos Alamos National Laboratory – Accelerated Load Cycle Durability Testing• R. Borup, R. Lujan, R. MukundanArgonne National Laboratory –Kinetic, rated power durability, and performance modeling.• R. Ahluwalia, X. Wang, J-K PengGeneral Motors - Stack Testing• B. Lakshmanan, N. Ramaswamy

19

Remaining Barriers

A. 2015(Sept.) Best of Class MEA does not achieve the DOE 2020 loading and specificpower targets.

B. Operational robustness of enhanced durability “B” interlayer is not maintained after30k Electrocatalyst AST cycles.

C. 2015(Sept.) BOC MEA is not sufficiently durable to achieve MEA load cycle durabilitytargets (10% V loss after 5000 hours).

1. Dealloyed Pt3Ni7/NSTF cyclic durability insufficient2. PFSA decomposition leading to cathode deactivation.

D. Operational robustness of 2015(Sept.) BOC MEA has not been demonstrated to beacceptable for automotive traction applications.

E. Rapid break-in conditioning method not yet implemented w/ BOC MEAs.

20

Key Future Work – FY16 (Through Aug. ‘16)

A. Complete short stack testing to evaluate operational robustness of projectBOC MEAs in relevant architecture.

1. Determine method to enable rapid break-in conditioning of BOC MEAs,compatible with stack testing.

2. Evaluate stack towards performance, cold/freeze startup, load transient.

B. Complete evaluations of downselected Best of Class MEA:1. ANL performance model2. Load cycle durability and AST durability.

C. Improve load cycle durability by integration of higher durability NSTFcathodes and experimental PEMs with reduced degradation contaminantimpact.

D. Complete assessment of relative cost savings of project Best of Class MEA tobaseline.

E. Complete project final report.

21

SummaryOperational Robustness (Cold Start; Load Transient)• Integrated new anode GDL and improved durability “B” cathode interlayer into 2015 (Sept.)

BOC MEA, yielding 0.1-0.2V improvement at intermediate T over 2015(Mar.) BOC MEA.

Durability (MEA Load Cycling; Electrocatalyst/Support ASTs)• New “B” interlayer maintains operational robustness after Support AST and after 20k

Electrocatalyst AST cycles. H2/Air performance near 1.5A/cm2 improves after ASTs.• Load cycle evaluation of BOC MEAs show significant reversible and irreversible

performance losses after 1200+ hours; tests ongoing.• New “M” NSTF cathode passes DOE Electrocatalyst AST, but improved activity needed.

Power, Cost (Cathode Post Processing; Best of Class MEA Integration)• Substantial gains in specific power (55% kW/g v. pre-proj.) due to improved absolute

performance and PGM reduction. DOE 2020 targets for loading, rated power approached.• Pilot scale components successfully fabricated.

Short Stack Evaluation of BOC MEAs• To date, BOC MEA evaluated in GM short stacks have not demonstrated expected

performance and robustness benefit over baseline. • Insufficient conditioning of ultra-low PGM anode suspected• Development of stack-compatible, rapid conditioning methods are in progress.

22

Technical Back-Up Slides

23

Project Goal TableTable 10. Performance, Cost, Durability Targets, Current Project Status, and Go/No-Go and Goal Criteria

Performance at ¼ Power, Performance at rated power, and Q/∆T TargetsGoalID Project Goals (units)

Target Value

Status (NEW since ‘15

AMR)

G/NG or Interim

Goal Value1

Performance at 0.80V (A/cm2); single cell, ≥80ºC cell temperature, 50,100,150kPag, respectively.

0.300NANA

0.310 A

NANA

0.250 ≥0.300 ≥0.300

2 Performance at Rated Power, Q/∆T : Cell voltage at 1.41A/cm2

(Volts); single cell, ≥88ºC cell temperature, 50kPag* 0.709 0.679A 0.659

Cost Targets3 Anode, Cathode Electrode PGM Content (mg/cm2) ≤ 0.125 0.131A 0.1354 PEM Ionomer Content (effective ion. thickness, microns) ≤ 16 12A 20

Transient response (time from 10% to 90% of rated power), Cold start up time to 50% of rated power at -20°C, +20°C), and Unassisted start.

5 Transient response (time from 10% to 90% of rated power); single cell at 50°C, 100% RH (seconds)

≤ 1PASS

(0%RH)A 5

6 Cold start up time to 50% of rated power at +20°C; evaluated as single cell steady state J at 30°C (A/cm2)

≥ 0.8 0.7A 0.6

7 Cold start up time … at -20°C; short stack (seconds) ≤ 30 27C 308 Unassisted start from -40°C (pass/fail); short stack Pass at -

40°CPass at -20°CC

Pass at -30°C

MEA Durability with cycling, Electrocatalyst Cycle, Catalyst Support Cycle, MEA Chemical Stability, and Membrane Mechanical Targets

9 Cycling time under 80°C MEA/Stack Durability Protocol with ≤ 30mV Irreversible Performance Loss (hours)

≥ 5000 TBDA 2500

10 Table D-1 Electrocatalyst Cycle and Metrics (Mass activity % loss; mV loss at 0.8A/cm2; % initial area loss)

≤-40≤-30 ≤-40

-40±0.8-28±1.4

-14±0.2 F

≤-40≤- 30 ≤-40

11 Table D-2 Catalyst Support Cycle and Metrics (Mass activity % loss; mV loss at 1.5A/cm2; % initial area loss)

≤-40≤-30≤-40

-40±7-11±3 (0.8)

-19±3E

≤-40≤-30≤-40

12 Table D-3 MEA Chemical Stability: 500 hours (H2 crossover (mA/cm2); OCV loss (% Volts); Shorting resistance (ohm-cm2))

≤2≤-20

>1000

PASS-4

PASSB

≤2≤-20

>100013 Table D-4 Membrane Mechanical Cycle: 20k Cycles (H2

crossover (mA/cm2); Shorting resistance (ohm-cm2))≤2

>100020.1kA (PEM

ONLY)≤3

>500

A: Mean values for duplicate 3M 2015(Sept.) Best of Class NSTF MEAs: Anode=0.02PtCoMn/NSTF, Cathode= 0.095Pt3Ni7/NSTF + 0.016Pt/C IL, (0.131mPGM/cm2 total), 3M-S 725EW 14µ PEM, X2/2979+IL Anode/Cathode, 3M “FF2” flow fields, operated at 90ºC cell temperature with subsaturated inlet humidity and anode/cathode stoichs of 2.0/2.5 and at stated anode/cathode reactant outlet pressures, respectively.B: Mean value for duplicate 3M 2015(Mar.) Best of Class MEAs. Analogous result for 2015(Sept.) MEAs is expected.C: OEM Stack testing results with 3M NSTF MEAs: Anode=0.10PtCoMn/NSTF, Cathode=0.15PtCoMn/NSTF, (0.25mgPGM/cm2 total), 3M ionomer in supported PEM, Baseline 2979/2979 GDLs. OEM-specific enabling technology.E: Value for Replicate 3M NSTF MEAs. Anode: 0.05PtCoMn/NSTF. Cathode=0.107 or 0.125 Pt3Ni7/ NSTF(Dealloy+SET), 3M 825EW 24µ PEM w/ or w/o additive, Baseline 2979/2979 GDLs, w/ or w/o Edge Protection, Quad Serpentine Flow Field.E: Value for Replicate 3M NSTF MEAs. Anode: 0.05PtCoMn/NSTF. Cathode: “M”, 0.091mgPGM/cm2, 3M 825EW 24µ PEM w/ or w/o additive, Baseline 2979/2979 GDLs, w/ or w/o Edge Protection, Quad Serpentine Flow Field.

*: Cell performance of 0.709V at 1.41A/cm2 with cell temperature of ≥88ºC simultaneously achieves the Q/∆T and rated power targets of 1.45kW/ºC and 1000mW/cm2, respectively.**: Single sample result. MEA failed prematurely due to experimental error.

24

2015 (Sept.) BOC MEA Performance v. Pressure

0.0000.0250.0500.0750.1000.5

0.60.70.80.91.0

0.0 0.5 1.0 1.5 2.00.0000.0250.0500.0750.1000.5

0.60.70.80.91.0

0.0 0.5 1.0 1.5 2.0

FC035397 606.RAW_EISFIT_POLCURVE Cell Voltage Measured FC035397 606.RAW_EISFIT_POLCURVE Cell AC Impedance Measured FC035424 563.RAW_EISFIT_POLCURVE Cell Voltage Measured FC035424 563.RAW_EISFIT_POLCURVE Cell AC Impedance Measured

Cell

V (V

olts

), HF

R (o

hm-c

m2 )

90/84/84oC1.5/1.5atmA (OUT) H2/Air

CS2/2.5GDS(2min/pt)

Flow field flooding


C:\Users\US314230\Documents\DOE14\Task6\BOC MEA Startup Lab Pilot-[Graph7]

90/80/80oC2.0/2.0atmA (OUT) H2/Air

CS2/2.5GDS(2min/pt)


J (A/cm2)

90/76/76oC2.5/2.5atmA (OUT) H2/Air

CS2/2.5GDS(2min/pt)


J (A/cm2)

92/87/87oC1.5/1.5atmA (OUT) H2/Air

CS2/2.5GDS(2min/pt)

0.0 0.1 0.2 0.3 0.4 0.50.750

0.775

0.800

0.825

0.850

84/84oC A/C DP

90oC Cell, GDS(2min/pt)1.5/1.5atmA (OUT) H2/Air, CS2/2.5

68/68oC A/C DP

Cell

V (V

olts

)

J (A/cm2)

FF Flooding MitigationBy Reduced RH @ Low J

0.64 0.66 0.68 0.706

7

8

9

10

0.64 0.66 0.68 0.70 0.720.8

0.9

1.0

1.1

1.2

1.3 90C, 1.5atmA 90C, 2.0atmA 90C, 2.5atmA

Cell V (Volts)

90oC Cell T., CS(2,100)/CS(2.5, 167), H2/Air76-84C A, C Dewpoints

Spec

ific

Powe

r (kW

/g)

Each curve is average of 2 MEAs, interpolated at spcific voltages

90C, 1.5atmA 90C, 2.0atmA 90C, 2.5atmA

Cell V (Volts)

Powe

r (W

/cm

2 )

Each curve is average of 2 MEAs, interpolated at spcific voltages

27

Technical Backup – Progression Over Project

2012 (

March)

2012 (

Sept)

2013 (

March)

2013 (

Dec)

2014 (

Mar)

2015 (

Jan)

2015 (

Mar)

2016 (

Sept.

)0

4

8

120.6

0.8

1.0

1.2

1.40.10

0.12

0.14

0.16

Target8kW/g

250200150

Spec

ific

Powe

r (k

W/g

PGM) @

0.6

7V P(kPa)

Target1W/cm2

250200

150

Rate

d Po

wer

W/c

m2 @

0.6

7V

P(kPa)

Target0.125mg/cm2

Tota

l PGM

(mg/

cm2 )

0.0 0.5 1.0 1.5 2.00.6

0.7

0.8

0.9

J (A/cm2)

90oC Cell T., 7.35/7.35psig H2/Air (OUT), CS(2,100)/CS(2.5, 167), GDS(120s/pt, High->low J Only). 2012 BOC: 84/84C A/C Dewpoint

2015 BOC: 84/84C Dewpt for J > 0.4. 68/68C Dewpt for J < 0.4

2015(Sept.) Best of Class NSTF MEA - 0.131mgPGM/cm2 TotalANODE: 0.020PtCoMn/NSTF.

CATHODE: 0.095Pt3Ni7/NSTF (DEALLOY) + 0.016Pt/C Interlayer.PEM: 3M-S 14µ 725EW. GDL: X3/2979.

2015(Sept.) BOC MEA0.133mgPGM/cm2

6.8kW/g @ 0.692V

2012 BOC MEA0.151mgPGM/cm2

4.4kW/g @ 0.692VCell

Volta

ge (V

olts

)

0 20 40 60 80 100-0.2

0.0

0.2

0.4

0.6

0.8 2015 (Sept.) BOC MEA

RobustnessTargets

2015 (Mar.) BOC MEA

Min

imum

Cel

l V D

urin

gLo

ad T

rans

ient

(Vol

ts)


2013 BaselineMEA

150/150kPaA CS2/2 H2/Air. 1A/cm2

60-80oC: 100%RH 30-40oC: 0% RH

With IL

26

NSTF Anode Activation

Wang et al., “Kinetics of Hydrogen Oxidation and Hydrogen Evolution Reactions…”, J. Electrochem. Soc. 160(3) F251 (2013)

ElectrodeEHOR(kJ/mol)

i0 (mA/cm2

Pt)0.05PtCoMn/NSTF, Partially activated 38.9 4890.05PtCoMn/NSTF, Fully activated 64.3 2727 ± 5630.003Pt/C (5wt%) NA 235-300

1. Early project work: little apparent impact with 0.02mgPt/cm2 PtCoMn/NSTF anode(all unactivated).

0.00 0.25 0.50 0.75 1.00 1.25 1.500.0

0.2

0.4

0.6

0.8

1.0 Anode PGM (mg/cm2)PtCoMn/NSTF

0.005 0.01 0.02 0.03 0.05

Cell

Volta

ge (V

olts

)

J (A/cm2)

75/70/70C, 0/0psig H2/Air, 800/1800SCCMPDS(0.85V->0.25V->0.85V, 0.05V/step, 10s/step)

0 2 4 6 8 1040

30

20

10

0

NFAL Only NFAL+RFAL

η HO

R (m

V)

SEF (cm2-Pt/cm2-planar)

pH2=1atm, 1.5A/cm2, 80C

0.0 0.5 1.0 1.5 2.0-0.10

-0.05

0.00

After AnodeConditioning

Before AnodeConditioning

J (A/cm2)

80oC Cell, 100% RH, 1.5atmA , 1750SCCM H2

GDS(2min/pt).IR-F

ree

Cell

V (V

olts

)

2. NSTF HOR kinetic study concluded that thermal cycle activation increased NSTF HOR specific activity > 5x over unactivated, and 1 order of magnitude over Pt/C.

3. Model prediction of ~10mV excess loss due to poor activation at 2.5cm2

Pt/cm2planar (0.02PtCoMn/NSTF)

4. Actual HOR kinetics much lower than model prediction, or by early project FC tests (#1)• 60mV before activation (v. 15mV modeled)• 15mV after activation ( v. 3mV modeled)

27

-- Unlimited Rights Data --

Short Stack Design

Following GDLs were used Anode – 3M-X2 Low phobic GDL Cathode – 3M 2979

Short stack stands were equipped with DI Water Flush lines for thermal cycle break-in process

# Type Anode Description

Membrane Cathode Description No. of Cells

1 3M Baseline CCM PtCoMn/NSTF0.05 mgPt/cm2

3M 825EW20 µm

PtCoMn/NSTF0.15 mgPt/cm2 4

23M ‘Baseline’ CCM

with 14 µm membrane

PtCoMn/NSTF0.05 mgPt/cm2

3M-S 725EW14 µm

w/ additive

PtCoMn/NSTF0.15 mgPt/cm2 4

3 3M March BOC CCM PtCoMn/NSTF0.019 mgPt/cm2

3M-S 725EW14 µm

w/ additive

Dealloyed Pt3Ni7/NSTF0.096 mgPt/cm2

3M 2979 w/ “B” Interlayer IL Loading – 0.016 mgPt/cm2

10

4 GM Baseline CCM Dispersed Pt/C800EW Ionomer 18 µm PFSA Dispersed Pt-alloy/C

800EW Ionomer 10

Stack consisted of the following CCM MEAs