Long Life PEM Water Electrolysis Stack Experience and Future Directions E. Anderson, K. Ayers, C. Capuano and S. Szymanski Technical Forum Hannover Messe Germany 9 April 2013 ™®

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.

Proton’s Markets, Products & Capabilities Power 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

Over 10 MW Shipped – Future Growth from MW-Scale PEM Electrolysis 2

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

Courtesy of Siemens AG, 2012 4

Sandia National Laboratory Analysis Cost analysis shows cost effectiveness of hydrogen RFCs Sandia public report: SAND20114845 Batteries

Electrolysis

Storage  Expandable with   Capacity > 99% reliability

Corrective actions in place to address top 2 failure modes

9

Established PEM Stack Durability 3.0

Proton Energy Systems In-House Cell Stack Endurance Testing

Average Cell Potential (Volts, 50o C)

200 psig (13 barg) 2.6 2003 Stack Design: 1.3 A/cm2 4 µV/cell hr Decay Rate 2.2

2005 Stack Design: 1.6 A/cm2 Non Detectable Decay Rate

1.8

1.4 0

10,000

20,000

30,000

40,000

50,000

60,000

Operating Time (Hours)

~60,000 hour life demonstrated in commercial stack designs New designs have no detectable voltage decay 10

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 U.S. Funding Agencies: NSF ARPA- E DOE-EERE

ONR

CERL

TARDEC

Development Stage/Risk Level Materials Feasibility Applied research Demonstration R&D

Deployable Prototypes

11

PEM Cost and Efficiency Limitations • Flow field, membrane electrode assembly, and labor are high impact cost areas • Efficiency losses dominated by membrane ionic resistance and O2 reaction overpotential

Electronics

Cell stacks

System vs. stack breakdown

12

Technology Roadmap Execution • Clearly defined pathways enable directed research Cell Stack Technology Roadmap 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

Page 13

Manufacturing Development for High Volume and Automation

Production Implemented

Funded Not Yet Funded

Cost Reduction Initiatives Noble Metal Reduction 2.2 2

Potential (Volts)

1.8 1.6 3M Cathode #1, 8 mil total 3M Cathode #2, 8 mil total Proton Baseline , 7 mil

1.4 1.2 1 0.0

0.2

0.4

0.6 0.8 1.0 1.2 1.4 1.6 Current Density (Amps/cm2)

1.8

2.0

2.7 2.6 2.5

Approximate current production range

2.4 2.3 Potential (V)

3M NSTF electrode: 5% of current catalyst loading

Flow Field Cost

2.2 2.1

Adv membrane: Enables 2X current (lower Capex)

2 1.9 1.8 1.7

Reduced voltage (lower Opex)

1.6 1.5

90 micron 60 micron

1.4 0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

5.5

6

Current Density (A/cm2 )

Lower Cost Membranes 14

New design with ~50% less metal

Efficiency Needs and Progress: Membrane • Legacy membrane does not hold up at desired thicknesses and pressures 2.05 1.95

Voltage (V)

• Optimization of Tg, IEC, and reinforcement show good mechanicial durability

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

1.85 1.75 1.65 1.55 0

Page 15

2000

4000 6000 Time (h)

8000

10000

Cost Needs and Progress - Catalyst • Engineered structures for ultra-low PGM loadings 3M NSTF

2.2

Potential (Volts)

2 1.8 1.6 3M Cathode #1, 8 mil total 3M Cathode #2, 8 mil total Proton Baseline , 7 mil

1.4 1.2

Brookhaven GDE approach

1 0.0

0.2

0.4

0.6 0.8 1.0 1.2 1.4 1.6 Current Density (Amps/cm2)

1.8

2.0

Polarization curves: 3M loading < 1/20th baseline, Brookhaven 20,000 hours validated on 3-cell – > 1,500 hours on 10-cell stack – Full-scale +50 kg/day stack scale-up in process 2.40 Large Area Cell Short Stack 1032 amps, 30 barg, 50°C

Avg Cell Potential (V)

2.30 2.20 2.10 2.00 1.90 1.80 1.70 0

5000

10000

15000

20000

25000

Operating Hours

21

Proton Stack Performance Progression Technology Progression 2.4 Baseline,  50C Advanced Oxygen Catalyst, 50C Advanced membrane,  80C Advanced cell design,  80C

2.3 2.2

Current Stack (~70% Eff (HHV)

Potential (Volts)

2.1 2 1.9 1.8 1.7

Advanced Stack (>86% Eff (HHV)

1.6 1.5 1.4 0

0.5

1

1.5

2

2.5

Current Density (A/cm2)

• Additional testing at current densities >5 A/cm2 22

3

H2 Generation Capacity Tradeoffs Cell Potential  (Volts)

• Stack capable of current density • Evaluate impact on BoP ratings − Power buss − Thermal control − Gas drying • Added cost of lower utilization extra capacity?

2.4

80˚C, 30 bar H2 Pressure

2.3 2.2 2.1 2 1.9 1.8 1.7

Current Generation  MEA Next‐Gen MEA1 Next‐Gen MEA2

1.6 1.5 1.4 0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Current Density (A/cm2)

Power supplies 32%

Balance of plant

48% 25% 53%

MEA

5% 3% 13% 12% 24% 23%

flow fields and separators balance of cell

15%

balance of stack

Stack System 23

4.0

4.5

5.0

5.5

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

Large Active Area PEM Stack

– Transition from previous fully-packaged designs

• Trade-off Capex vs. efficiency for specific market applications 24 - Proton OnSite Proprietary and Confidential -

MW-scale concept

Scale-Up Cost Reduction Trajectory

Percent Cost Reduction

100%

80%

60%

40%

20%

0% C30

0.5 MW

1 MW

2 MW

2 MW Adv

Electrolysis System Size

• Straight-forward engineering scale-up of current products • Critical technology elements already developed 25 - Proton OnSite Proprietary and Confidential -

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 26