™ ® Proton, Proton OnSite, Proton Energy Systems, the Proton design, StableFlow, StableFlow Hydrogen Control System and design, HOGEN, and FuelGen are trademarks or registered trademarks of Proton Energy Systems, Inc.

Any other brands and/or names used herein are the property of their respective owners.

Long Life PEM Water Electrolysis Stack Experience and Future Directions

E. Anderson, K. Ayers, C. Capuano and S. Szymanski

Technical ForumHannover Messe

Germany9 April 2013

Proton’s Markets, Products & CapabilitiesPower Plants

Heat Treating

Semiconductors

Laboratories

Government

• Complete product development, manufacturing & testing

• Turnkey product installation• World-wide sales and service• Containerization and hydrogen

storage solutions• Integration of electrolysis into RFC

systems

2

Over 10 MW Shipped – Future Growth from MW-Scale PEM Electrolysis

Critical Needs for Energy Storage

• Renewable energy is growing rapidly world-wide in both wind and solar– Inherent intermittency has more impact as RE becomes a

larger portion of the grid capacity

– Up to 20-40% of wind energy can be stranded without storage

• Need generation technologies for storing excess renewable capacity & balancing loads on the grid

• Energy storage can also provide a linkage between utilities & transportation

3

Energy Storage Segmentation Map

4

Courtesy of Siemens AG, 2012

Sandia National Laboratory Analysis

Batteries Electrolysis

Storage Capacity <10 hours

Expandable with  inexpensive storage

Cycle Life Limited UnlimitedCost Challenges Yes

Leverage fuel cell scale up

Scale Unproven 100 year technology

5

Cost analysis shows cost effectiveness of hydrogen RFCs

Sandia public report: SAND2011-4845

Markets for MW Electrolysis• Transportation market

– H2 infrastructure plans in Europe, Asia• Renewable energy storage

– 10’s of GW’s of wind/solar energy capture– Power-to-gas

• Biogas market– Methanization

• Multi-billion dollar opportunity in each• MW-scale electrolysis needed

– Targeting 1-2 MW product scale-up

6

H2 Energy Product Attributes• Reliability – Cannot afford down-time or high replacement

costs

• Cost – Price targets for energy market apps more difficult (Capex vs. Opex)

• Load-following – Need to handle fast response of varying renewable power

• Operating Range – Need to handle wide variability of available renewable power(i.e. 100% turndown)

• Efficiency – Cost of electricity impact (Opex)

• Scale – MW-class electrolysis required

7

Today…

• Proton’s PEM-based hydrogen generators are demonstrating excellent reliability in industrial applications

• Cell stack technology is the most reliable component in the system

• New energy applications present capital as well as operating cost challenges

• Necessary technology advances are the biggest risk to established durability and reliability

8

Cell Stack Reliability – Over the past 6 yrs…

9

Corrective actions in place to address top 2 failure modes

Hundreds of millions of cell-hours of field experience>> 99% reliability

Established PEM Stack Durability

10

~60,000 hour life demonstrated in commercial stack designsNew designs have no detectable voltage decay

1.4

1.8

2.2

2.6

3.0

0 10,000 20,000 30,000 40,000 50,000 60,000

Average Cell Potential

(Volts, 50oC)

Operating Time (Hours)

2003 Stack Design:1.3 A/cm2

4 µV/cell hr Decay Rate

200 psig (13 barg)

Proton Energy SystemsIn-House Cell Stack Endurance Testing

2005 Stack Design:1.6 A/cm2

Non Detectable Decay Rate

11

Technology Roadmap Development

• Clearly defined pathways enable directed research– Product, cell stack, and balance-of-plant

• Balance portfolio with near and long term R&D• Leverage 3rd party funding to subsidize internal R&D

– Utilize military and aerospace as early adopters– Develop key partnerships to broaden skill base

• Feed into commercial markets as proven

NSF ARPA- E DOE-EERE ONR CERL TARDEC

Materials Feasibility Applied Deployable Prototypesresearch Demonstration R&D

U.S. Funding Agencies:

Development Stage/Risk Level

PEM Cost and Efficiency Limitations

12

• Flow field, membrane electrode assembly, and labor are high impact cost areas

• Efficiency losses dominated by membrane ionic resistance and O2 reaction overpotential

System vs. stack breakdown

Cell stacksElectronics

Technology Roadmap Execution• Clearly defined pathways enable directed research

Page 13

0-18 months 1.5-3 years 3-5 years

Catalyst Cost & Efficiency: Loading Reductions (Process and Structure) and Composition Optimization

Thinner Membranes: Improved Mechanical and Thermal Stability and Reduced Crossover

Flow Field Material and Coatings for H2/O2 protection

Scale Up in Capacity and Pressure

Manufacturing Development for High Volume and Automation

Cell Stack Technology Roadmap ProductionImplemented

Funded

Not Yet Funded

Cost Reduction InitiativesNoble Metal Reduction Flow Field Cost3M NSTF electrode:5% of current catalyst loading

New design with ~50% less metal

14

1

1.2

1.4

1.6

1.8

2

2.2

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

Potential (Vo

lts)

Current Density (Amps/cm2)

3M Cathode #1, 8 mil total3M Cathode #2, 8 mil totalProton Baseline , 7 mil

1.4

1.5

1.6

1.7

1.8

1.9

2

2.1

2.2

2.3

2.4

2.5

2.6

2.7

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6

Potential (V)

Current Density (A/cm2)

Approximate current production range

90 micron

60 micron

Adv membrane: Enables 2X current(lower Capex)

Reduced voltage(lower Opex)

Lower Cost Membranes

Efficiency Needs and Progress: Membrane

• Legacy membrane does not hold up at desired thicknesses and pressures

• Optimization of Tg, IEC, and reinforcement show good mechanicialdurability

Page 15

1.55

1.65

1.75

1.85

1.95

2.05

0 2000 4000 6000 8000 10000

Volta

ge (V

)

Time (h)

2mil high Tg membrane(160 amps, 200 psi, 80°C)

Cost Needs and Progress - Catalyst

• Engineered structures for ultra-low PGM loadings

16

1

1.2

1.4

1.6

1.8

2

2.2

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

Potential (Vo

lts)

Current Density (Amps/cm2)

3M Cathode #1, 8 mil total3M Cathode #2, 8 mil totalProton Baseline , 7 mil

Polarization curves: 3M loading < 1/20th baseline, Brookhaven <1/30th loading

3M NSTF

Brookhaven GDE approach

Cost Needs and Progress: Flow Fields• New bipolar cell assembly design: 50% metal reduction

• Alternative coatings: Eliminates process stepsand mitigates hydrogenembrittlement

17

Flowfield/Bipolar Plate Reliability• Typical materials are semi-

precious metals– Precious metal coatings

added to reduce resistance• Susceptible to oxidation

(anode) or embrittlement (cathode) with prolonged operation

• Need lower cost alternatives• Investigating impact of

process methods & alternative non-precious metal coatings on durability

18

0%

20%

40%

60%

80%

100%

0 200 400 600 800Hyd

rogen Uptake vs. Baseline

Time (h)

Baseline

Treatment 1

Treatment 2

Nitrided

Accelerated H2 Exposure Testing(100 hrs 10 yr operation)

Intermittency and Variable Load Input• Electrolysis is well suited to load following

– Stable performance– Rapid response time to current signal– Tolerance to variable power input

19

Courtesy of NREL

50 ms response time demonstrated

Electrolysis Stack & System Development

20

650 cm2

30 Nm3/hr65 kg/day

1.70

1.80

1.90

2.00

2.10

2.20

2.30

2.40

0 5000 10000 15000 20000 25000

Avg

Cel

l Pot

entia

l (V

)

Operating Hours

Large Area Cell Short Stack1032 amps, 30 barg, 50°C

600 cm2 Stack Development• Improvement in bipolar plate design

– Current 86 cm2 design tested to over 1 million cell hours

– CFD modeling shows more uniform flow• Demonstrated operation up to 30 bar

– >20,000 hours validated on 3-cell– > 1,500 hours on 10-cell stack– Full-scale +50 kg/day stack scale-up in process

21

Proton Stack Performance Progression

• Additional testing at current densities >5 A/cm2

22

1.4

1.5

1.6

1.7

1.8

1.9

2

2.1

2.2

2.3

2.4

0 0.5 1 1.5 2 2.5 3

Potential (Vo

lts)

Current Density (A/cm2)

Technology Progression

Baseline, 50CAdvanced Oxygen Catalyst, 50CAdvanced membrane, 80CAdvanced cell design, 80C

Current Stack(~70% Eff (HHV)

Advanced Stack (>86% Eff (HHV)

H2 Generation Capacity Tradeoffs• Stack capable of current density• Evaluate impact on BoP ratings

− Power buss− Thermal control− Gas drying

• Added cost of lower utilization extra capacity?

23

15%

32%

13%

25%3%

12%53%

Power supplies

Balance of plant

MEA

flow fields and separators

balance of cell

balance of stack

24%

48%

5%

23%

StackSystem

1.4

1.5

1.6

1.7

1.8

1.9

2

2.1

2.2

2.3

2.4

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5

Cell Po

tential (Vo

lts)

Current Density (A/cm2)

80˚C, 30 bar H2 Pressure

Current Generation MEANext‐Gen MEA1Next‐Gen MEA2

MW-Scale Product Development Pathway

• Multi-stack architecture– Consistent with H- and C-Series products

• Large active area stack platform– 3X Increase over C-Series stack– Prototype already developed and tested– Advanced design to deliver 40% cost reduction

• Open-frame skid modular plant configuration– Transition from previous

fully-packaged designs• Trade-off Capex vs. efficiency

for specific market applications

Large Active Area PEM

Stack

MW-scale concept

24- Proton OnSite Proprietary and Confidential -

Scale-Up Cost Reduction Trajectory

• Straight-forward engineering scale-up of current products• Critical technology elements already developed

0%

20%

40%

60%

80%

100%

C30 0.5 MW 1 MW 2 MW 2 MW Adv

Perc

ent C

ost R

educ

tion

Electrolysis System Size

25- Proton OnSite Proprietary and Confidential -

26

Summary

• Industrial PEM electrolysis systems have excellent reliability track record

• New energy applications will challenge that reliability as technology advancements to drive cost and reliability are adopted

• Technology roadmap is guiding progress and funding of continued development

• Market needs for hydrogen energy storage are emerging rapidly

• Proton is a leader in this technology space