pem & alkaline electrolyzers bottom-up manufacturing cost … cell/yang_pem and alkaline...

Austin Power Engineering LLC1 Cameron St

Wellesley, MA 02482USA

[email protected]

© 2017 Austin Power Engineering LLC

PEM & Alkaline Electrolyzers Bottom-up Manufacturing Cost Analysis

Yong Yang

Austin Power

November 8, 2017

David Hart

E4tech

http://www.austinpowereng.com/

12017 YY/DH

Water electrolysis could play a significant role in the future energy system …

Introduction Market Overview

Roadmap for electrolyser deployment by application. Source: 2014 FCH-JU study on water electrolysis

Consolidated views on rollout potential

Germany is considering to support roll-out of ~1GW

of electrolysers in the 2020ies timeframe

Multi MW demo projects in industry

A 2017 study commissioned by the FCHJU found that by 2025, an estimated cumulative electrolyser capacity of 2.8 GW could be installed in Europe

0

500

1,000

1,500

2,000

2,500

3,000

2010 2015 2020 2025 2030

[EUR/kW] Capital cost for Alkaline systems

Alkaline (all data)Central caseRange

0

500

1,000

1,500

2,000

2,500

3,000

2010 2015 2020 2025 2030

[EUR/kW] Capital cost for PEM systems

PEM (all data)Central caseRange

22017 YY/DH

… if predicted cost reductions can be realised.


Data source cost reduction expectations: 2014 FCH-JU study on water electrolysis

Expert ‘predicted’ cost reductions

Prices of ~500EUR/kW are now being quoted for

50MW+ systems

(MW-scale) (MW-scale)

Prices of

32017 YY/DH

Today’s electrolyser market is comparatively small, the industry landscape is highly fragmented and no single design dominates


• Perhaps 4% of global hydrogen supply is

produced via electrolysis (incl. Chlor-Alkali)

• Larger (> 50 kW) water electrolysis systems

are typically deployed for continuous operation.

Examples are fertilisers and methanol

production, fats & oils, float glass, …

• The industry is highly fragmented, with a few

established players and many start-ups / new

entrants – and cost reduction potential

18%

4%

48%

30%

Coal gasification

Electrolysis

Natural gas reforming

Refinery/Chemical off-gases

18%

4%

48%

30%

Coal gasification

Electrolysis



Source: IEA (2007)

18%

4%

48%

30%

Coal gasification

Electrolysis



18%

4%

48%

30%

Coal gasification

Electrolysis



Source: IEA (2007)

Chemistry Manufacturing Capacity

(MW/year)

Revenue

(million USD/year)

Alkaline ~300 MW ~100

PEM ~ 50 MW ~50

Total ~350 MW ~150

Rough estimates – revenue varies year-

on-year dependent on individual projects.

Current manufacturing capacities are

underutilised, but could be scaled up

quickly if demand rises.

We work with clients with various electrolysis technologies from the technology development stage to the large scale (GW) deployment.

42017 YY/DH

Technology

Development

Technology in Early

Commercial Stage

Mature

Products

Demonstration

Projects

Large Scale

Deployment

PEM

SOEC

Alkaline

Future

AlkalineFuture

PEM

Introduction Our Clients

5

Approach Manufacturing Cost Modeling Methodology

This approach has been used successfully for estimating the cost of various technologies for commercial clients and the DOE.

Technology

Assessment

Manufacturing

Cost Model

Scenario

Analyses

Verification &

Validation

• Literature research• Definition of system and component diagrams• Size components• Develop bill-of-materials (BOM)

• Define system value chain• Quote off-shelve parts and materials• Select materials• Develop processes• Assembly bottom-up cost model •Develop baseline costs

• Technology scenarios• Sensitivity analysis • Economies of Scale• Supply chain & manufacturing system optimization• Life cycle cost analysis

• Cost model internal verification reviews• Discussion with technical developers• Presentations to project and industrial partners• Audition by independent reviewers

Membrane

8.0%

Stack

Conditioning

2.7%

Seal

8.4%

Balance of Stack

2.4%

Bipolar Plate

26.1%

GDL

5.1%

Electrode

41.0%

Stack Assembly

6.3%

2017 YY/DH

Anode Ink

Anode Side

Catalyst Layer

Membrane

Processes

Cathode Side

Catalyst Layer

Cathode Ink

Hot Press

Lamination

Purchased GDL

Purchased GDL

Pt

Pt

Nafion®

Ionomer

99.9% 98.5%

99.9% 98.5%

99% 98.5%

100% 98.5%99% 98.5%

99% 98.5%

100% 100%

100% 100%

100% 100%

100% 100%

100% 100%

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

3-Year TCO 5-Year TCO 10-Year TCO 15-Year TCO

To

tal

Co

st

of

Ow

ers

hip

($

)

PEMFC Plug Hybrid Vehicle

Full Battery Electric Vehicle

6

Combining performance and cost model will easily generate cost results, even when varying the design inputs.

2017 YY/DH

Process Simulation

Energy requirements

Equipment size/ specs

Product Costs

Product cost (capital,

O&M, etc.)

Conceptual Design

System layout and

equipment requirements

Capital Cost EstimatesSite Plans

Safety equipment, site

prep, land costs

High and low volume

equipment costs

Air (POX only)

Nat. Gas

Water

Fuel

Reformer PSA

H2-rich gas

H2-poor gas

Catalytic

Burner

HeatCold

Water

99.99% pure H2

Low

Pressure

Storage

Medium

Pressure

Storage

High

Pressure

Storage

Flow

cntrlr

Flow

cntrlrFlow

cntrlr

Dispenser

To Vehicle

CO2

H2O

Compressor with intercoolers

Cooling

Tower

Process cost

Material cost

Tape Cast

Anode

Powder Prep

Vacuum

Plasma

Spray

Electrolyte

Small Powder

Prep

Screen

Print

Cathode

Small Powder

Prep

Sinter in Air

1400CSinter in Air

Forming

of

Interconnect

Shear

Interconnect

Vacuum

Plasma

Spray

Slurry

Spray

Screen

Print

Slurry Spray

Slip Cast

Finish Edges

Note: Alternative production processes appear in gray to the

bottom of actual production processes assumed

Braze

Paint Braze

onto

Interconnect

Blanking /

Slicing

QC Leak

Check

Interconnect

Fabrication

Electrolyte CathodeAnode

Stack Assembly

Fuel Station Perimeter

Electrolyzer or SMR,

High-Pressure

Compressor

H2 High Pressure

Cascade Storage

System

Gaseous Fuel

Dispensing Islands

Underground Piping with shared conduit

Vent

Building

Covered Fueling Island

CNG High Pressure

Cascade Storage System

Fire Detector

Walls with minimum fire rating of 2 hours

Wall openings (doors & windows)

Building intake vents

Adjoining property that can be built on

Public sidewalks, parked vehicles

Streets

Railroad tracks

Other flammable gas storage

5 ft

25 ft

50 ft

10 ft

15 ft

10 ft

15 ft

10 ft

Minimum set-back distances for hydrogen and/or natural gas

storage (more stringent gas noted if relevant)

Property of:

TIAX LLC

1061 De Anza Blvd.

Cupertino, CA 95014

Task 5 CNG/Hydrogen Fueling

Site Plan - Fueling Station

Hydrogen and CNG fueling station

SIZE DWG BY DWG NO REV

A Stefan Unnasch B0228 - S0022 1

SCALE 1" = 8 ft 5 Jan 2004 SHEET 1 OF 110 ft

Security FenceNG line in

ft

ft

ft

ft

ft

ft

ft

ft

Process Cost Calcs

Material

Process

$0

$1

$2

$3

$4

$5

$6

$7

$8

$9

$10

Lapp

ing

CM

P

Wet

ther

mal

Oxida

tion

Spin

Coa

ting

Wet

Etc

hing

SiO

2

DC S

putte

ring

Ni

RF

Sput

terin

g E

Spin

Coa

ting

Step

per

DC S

putte

ring

Ag

Strip

ping

Spin

Coa

ting

Step

per

Dry

Etc

hing

SiO

2

DRIE

Si

Wet

Etc

hing

SiO

2

Dry

Etc

hing

NI

Wafe

r C

ost

($)

Approach Manufacturing Cost Modeling Methodology

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

3-Year TCO 5-Year TCO 10-Year TCO 15-Year TCO

To

tal

Co

st

of

Ow

ers

hip

($

)

PEMFC Plug Hybrid Vehicle

Full Battery Electric Vehicle

7

Austin Power Engineering’s manufacturing cost models can be used to determine a fully loaded selling price to consumers at high or low volumes.

DirectLabor

DirectMaterials

FactoryExpense

GeneralExpense

SalesExpense

Profit

Consumer

Selling Price

Fixed Costs

• Equipment and Plant

Depreciation

• Tooling Amortization

• Equipment Maintenance

• Utilities

• Building

• Indirect Labor

• Cost of capital

• Overhead Labor

Variable Costs

• Manufactured Materials

• Purchased Materials

• Fabrication Labor

• Assembly Labor

• Indirect Materials

Corporate Expenses

• Research and Development

• Sales and Marketing

• General & Administration

• Warranty

• Taxes

Approach Manufacturing Cost Structure

Manufacturing

Cost

We assume 100% financing with an annual discount rate of 10%, a 10-year equipment life, a 25-year building life, and three months working capital.

2012 YY

The hydrogen electrolysis system design is primarily driven by the output hydrogen flow rate, purity, and pressure requirements.

System Analysis

82017 YY/DH

Stack

Power

Supply

DI Water

O2/liquid

Separator

H2/liquid

Separator

KOH

Scrubber

H2 Boost

CompressorDe-Oxo H2 Dryer

Stack

Cooling

H2 Pre-

Cooler

H2 After

Cooler

Generic Hydrogen Electrolysis Plant Major Components

Not in PEM

based system

9

Different Alkaline system and stack designs can be tested in in different scenarios

Alkaline System Stack Specifications

2017 YY/DH

Alkaline Unit Current Future

System size MW

10

In Alkaline stacks, higher current densities are key to reducing costs, but all options are subject to fundamental limits.

Alkaline System Stack Cost Analysis

2017 YY/DH

Cost contributors Pathways to cost reduction Identified limits

Membrane and

electrodes

Increase current density and

reduce materials use through

innovative component design and

advanced electrode materials

Industry reports 0.6 to 0.8 A/cm² as

a long term target. Limitation is

acceptable efficiency at high

currents. (Current density today

typically: 0.2 A/cm²)

Cell frame Reduce material use through

large cell concepts with a better

ratio between active area and

frame area

Mechanical stability of cell

components, depending on stack

pressure level

Cell frame material Replace by (cheaper) injection

moulded thermal plastic, though

mechanical stability at pressure

may be limited

Mechanical stability of thermal

plastic

11

Different PEM system and stack characteristics for current and future designs can be tested in in different scenarios

PEM System Stack Specifications

2017 YY/DH

PEM Unit Current Future

System size MW 1

Current density A/cm² 1 ~2 > 2

Membrane material Nafion 200µm Nafion 200µm

Catalyst Ir, Pt Ir, Pt

Catalyst loading mg/cm² ~ 5 < 1

Conductive porous layer Ti foam Ti foam

Screen pack plates Ti meshFunction replaced by flow field

plates

Bipolar plate Ti foil Ti foil; or SS316 with Ti coating

Cell frame material Ti or polymer based Polymer based with metal inlay

12

PEM cost reduction opportunities are strongly related to the reduced use of high price materials, though other components contribute.

PEM System Cost Analysis

2017 YY/DH

Cost contributors Pathways to cost reduction Identified limits

Components made of

titanium

(cell frame, screen pack,

porous plate)

Reduce use of titanium in

components. Make bi-polar

plates of steel plus titanium

coating instead of full titanium.

Increase current density.

Titanium (high material and

processing cost)

expected to remain as the

material of choice for ‘acidic’

electrolysers

Electrode catalysts Material changes using

advanced catalyst support

structures, mixed metal oxides

and nano-structured catalysts.

Precious metal cost and

catalyst activity for acceptable

efficiency. Goals are:

- 0.3mg/cm² Iridium - Anode

- 0.1mg/cm² Pt - Cathode

Supplied membrane High volume orders and/or

dedicated production

Costly fluorine chemistry in

Nafion production

13

At the GW scale electrolysis plant, centralized BOP system might lead to a lower system cost.

System BOP Cost Analysis

2017 YY/DH

Cost contributors Pathways to cost reduction

Power supply 100MW+ electrolyzers mean lower engineering cost for

each individual rectifier. ~20 MW rectifier units are

common in other industries (aluminium smelters, chlor-

alkali electrolysis)

Gas/liquid separator

/ KOH Scrubber

Integrated KOH scrubber with gas/liquid separator will

reduce the overall cost

H2 booster compressor Pressurized stack design will eliminate the H2 booster

compressor

Deoxo unit Select low temperature catalyst will help reduce the vessel

material cost as well as reduce the H2 gas cooling unit

cost. Differential pressure PEM design can eliminate need

for de-oxo unit

H2 dryer Thermal swing adsorption dryer will have higher H2 gas

output (not consume H2 in regeneration step) which is

more important in the large scale applications

14

‘Expert-predicted’ cost reductions seem plausible and could make electrolysis much more competitive

Conclusions

2017 YY/DH

• Alkaline system and PEM system have broadly similar system costs at the high capacity

applications

• Alkaline cost reduction is quite sensitive to volume production (reduce the process

costs) and increased current density

• PEM routes to cost reduction include volume production, system scale-up,

reduction of expensive materials, and increased current density

• PEM could ultimately be lower cost than Alkaline, but PEM is less mature so future cost

is sensitive to a number of uncertain technology development assumptions

• BoP costs (mainly power supply) start to dominate system cost at high production

volumes, when other cost components have been compressed already

• Other technology may be interesting to analyse, for example SOEC

15

Thank You!

2017 YY/DH

Contact: Yong YangAustin Power Engineering LLC

1 Cameron StWellesley MA 02482

USA

Tel: +1 [email protected]

www.austinpowereng.com

Contact: David Hart E4tech

Av. Juste-Olivier 2 1006 Lausanne

Switzerland

Tel: +41 21 331 [email protected]

www.e4tech.com
mailto:[email protected]:[email protected]:[email protected]://www.e4tech.com/

pem & alkaline electrolyzers bottom-up manufacturing cost … cell/yang_pem and alkaline...

Documents