Low Cost, Off-board Regeneration of

Sodium Borohydride

Ying Wu

Millennium Cell Incorporated

26 May 2004

2004 DOE Hydrogen, Fuel Cells & Infrastructure

Technologies Program Review

This presentation does not contain any proprietary or confidential information This presentation does not contain any proprietary or confidential information

2

ObjectivesObjectives

Development of a Reliable Regeneration Process of Sodium Borohydride(NaBH4) that meets DOE Cost Targets

New Contract

Signed Feb 19, 2004

Technical Approach is to Identify Electrolytic Processes which Reduce Cost

Direct Borate Reduction

High Efficiency Sodium Reduction

A Key Tool is to use Hydrogen Gas To Reduce Cell Voltage and ImproveRegeneration Efficiency.

3

BudgetBudget

$1,097,326

$1,500,000

FY’04 Funding As of End of

March’04:

FY’04 –’06

$ 900,000

$3,600,000

$4,500,000

$70,846MCEL&APCI

$ 283,381DOE funds

$ 354,227Total Project

4

Technical Barriers and TargetsTechnical Barriers and Targets

Barriers

C. Efficiency

G. Life Cycle and Efficiency Analyses

Q. Regeneration Processes for IrreversibleSystems

R. By-Product Removal

Applicable to Delivery and Off-Board Storage

Targets

Will Meet Long Term Cost Target!

2010– $1.5 / gallon of gasoline equivalent

Energy Efficiency and Fuel Cost areIntimately Related

Regenerate By-Products into NewHydrogen Gas Carrier

100%

38%

21% 4% 2% 1%

$40.00

$71.00

$200.00

$4.60$7.10$1.40

0%

20%

40%

60%

80%

100%

120%

$/k

g S

BH

-20

0

20

40

60

80

100

120

140

160

180

200

$/k

g H

2

Proj'd SBH cost ($/kg)

Current SBH cost ($/kg)

Eq/ H2 Cost ($/kg)

Est. Prdctn.

Cost SBH

($3.5/kgNa)

Retail SBH

($3.5/kgNa)

Schl.

Process

($1.0/kg Na)

Direct

H-assist SBH

($0.5/kg NaBO2)

Direct H-assist

SBH (recyc'd

NaBO2)

Minimum Cost

($0.05/kWh,

$0.80/kg H2)

5

System Storage Density : Gravimetric Projections

Hydrogen on DemandTM System (50-75 kW, 7.5 kg stored H2 )

System Storage Density : Gravimetric Projections

Hydrogen on DemandTM System (50-75 kW, 7.5 kg stored H2 )

2.5% 3.0%

4.3%

6.9%7.8%

0

10

20

30

40

50

60

70

80

90

100

1998 2000 2002 2004 2006 2008 2010 2012 2014 2016

Year

NaB

H 4 S

tore

d C

on

cen

trati

on

(wt%

)

0

2

4

6

8

10

12

14

16

18

20

Syste

m H

2 S

tora

ge (

wt%

)

Stored Conc System wt% DOE Targets

Pre-Natrium

Natrium

Need FC Water!

Next Generation

Prototypes

Note: Projected System H2 Storage is based on an average weight.

NaBH4 Has an Exceptional Combination of Volumetric and Gravimetric EnergyDensities, and More Room for Upward Growth than Most Other Storage Technology!

6

4 NaH

(+B(OCH3)3)

4 Na

(+2 H2)

4 NaCl

NaBH4

365 kJ4 NaOH

(+2 H2)

500 kJ

100 kJ

1880 kJ

NaBH4 Regeneration is Keys To Cost ReductionNaBH4 Regeneration is Keys To Cost Reduction

• Direct reaction route toNaBH4 is highly desirable.

• Only one step!

• More efficient Naprocess will significantlyimpact NaBH4

regeneration efficiency

Reaction Steps

CO2

CO2Cl2 or HCl

CH4

Steam

Reforming

NaClElectrolysis

BoraxRefining

Na Metal

H2

CH3OH

NaH

H3BO3 B(OCH3)3

NaBH4

3 CH3OH + 3 NaOH

3 NaOCH3

H2SO4 Na2SO4

4

1

Reduce # of Process StepsReduce # of Process Steps Eliminate Inefficiency in Critical StepsEliminate Inefficiency in Critical Steps

3200 kJ

7

Technical ApproachTechnical Approach

Identify a one-pot electrolytic synthesis of the borohydride anion

Improve efficiency/cost of the present synthesis at most costlystep: Identify a reduced energy electrolysis to sodium metal

Integrated intoSchlesinger Process

Separator material

B2O3 removal from cell

E0rev = 2.16 V

H-assisted NaBO2

Electrolysis

Integrated intoSchlesinger Process

Separator material

Efficiency

E0rev = 1.07 V

H-assisted molten

NaOH electrolysis

Replacement technology

Separator materialNaBH4 stability

Cathode “catalyst”

E0rev = 0.89 V

Direct H-assisted

Electrolysis

Issues

How toImplement

Std CellPotential

8

Project SafetyProject Safety

General Safety Procedures

Fume Hood

Removable PlexiglassShielding

Distant Operator Interface

Standard and RegularSafety Inspections

Specific Safety Procedures

Double Bubbler

All Stainless SteelManifolding

“Pop-Top” Pressure Relief

Aluminum Housed Mantles

9

Task 1 – Study of the NaBH4 synthesis alternatives

Deliverable - Summary report Interim report on experimental work

Task 2 – Evaluate 3 methods for lower cost electrolytic synthesis

Deliverable - Detailed experimental report

Select Best Pathway to Proceed

Task 3 – Preliminary Engineering and Economics Study

Deliverable - Preliminary report on engineering on economic assessment

Select Best Pathway to Proceed

Task 4 – Laboratory Prototype Demonstration Unit

Deliverable - Recommendation for future development

Task 5 – Ongoing Project Management and Reporting

Deliverable - Final project report

Timeline of Proposed TasksTimeline of Proposed Tasks

Q1 Q3Q2 Q4 Q8Q7Q6Q5 Q9 Q10 Q11 Q12

Nov 1, 2005 – Oct 31, 2006Nov 1, 2003 – Oct 31, 2004 Nov 1, 2004 – Oct 31, 2005

Task 1 Task 3Task 2 Task 4

10

Progress and Technical AccomplishmentsProgress and Technical Accomplishments

New Project Start-Up ItemsAlternate Pathway Analysis Completed

Installation of New NMR CapabilitiesCompleted

Air Products Duplicate Set Up In Progress

Electrolysis ResultsSodium Metal Generated at Cell Potentialof 1.2 V

Apparent Electrolytic Activity of SodiumMetaborate in Hydroxide Melt Observed

Hydride Transfer Catalyst Study Completed

Anode

Tube madefrom ”-

alumina

Na+

Cathode

NaOHAnolyte

NaOHCatholyte

Na+

Na+

H2

Frit

Reference ElectrodeNa/Na+ couple in ”-

alumina tube

11

Electrolytic Reduction of NaOH to Na MetalElectrolytic Reduction of NaOH to Na Metal

Current Industrial Processes Require 9.7 kWh/kg Na Produced

H2 Assisted Electrolysis Shown Below Required only 1.8 kWh/kg Na!

Target is 1.5 kWh/kg Na, or 85% Efficiency

Reduced cell voltage and improved efficiency is directly applicable toelectrosynthesis of NaBH4.

Results

Theoretical Cell Voltage at 360°C Is 1.07 V

Steady State Operating Voltage: 1.2 V

Voltage Efficiency: 89%

Current Efficiency: 78%

Total Electrolytic Efficiency Is 69%

Results

Switching from Nitrogen to Hydrogen at theCounter Electrode Shows:

No Change in Cell Performance

Big Change in Cell Voltage

-500

-400

-300

-200

-100

0

0 3000 6000 9000 12000 15000 18000

Time (seconds)

Cu

rren

t (m

A)

-1.5

-1.2

-0.9

-0.6

-0.3

0

Cell P

ote

nti

al

(vo

lts)

Current Cell Potential

0

0.5

1

1.5

2

2.5

0 20 40 60 80 100 120 140

Time (seconds)

Po

ten

tial

(vo

lts)

-250

-200

-150

-100

-50

0

50

100

150

200C

urr

en

t (m

illiam

ps)

H2 Counter Electrode Potential N2 Counter Electrode Potential

H2 Current N2 Current

12

Apparent Electrolytic Activity of Sodium

Metaborate in Hydroxide Melt

Apparent Electrolytic Activity of Sodium

Metaborate in Hydroxide Melt

Direct Electrolysis of Borates Lowers Number of Process Steps

AND Requires Less Voltage than Current Methods of Sodium Production

AND May Allow Direct Reprocessing of the Chemical Hydride

-8

-6

-4

-2

0

2

4

6

8

-3 -2 -1 0 1 2

Potential (volts)

Cu

rren

t (m

A,

No

Bo

rate

)

-80

-60

-40

-20

0

20

40

60

80

Cu

rren

t (m

A,

Bo

rate

)

No Borate Borate

Results

Addition of NaBO2 to aHydroxide Melt CausesRadical Change in the CyclicVoltammogram

Reduction Wave RemainsUnidentified

13

Air Products and Chemicals IncorporatedSubcontractor

Engineering Expertise, Process Development, Scale-Up

Monthly In-person Project Team Meetings

Air Products Equipment on Loan to Millennium Cell

Princeton UniversityConsultant

Electrochemistry Expertise

INEELPreliminary discussion with Bruce Wilding on NaBH4 generation byradiation chemistry.

Ionotec LtdKey Supplier of Membranes

Technical Support, Custom Electrode Designs

Interactions and CollaborationsInteractions and Collaborations

14

Future Work(Year 1 of Project)

Future Work(Year 1 of Project)

Ascertain NaBO2 Electrolysis Results; Rule Out False Positives;Demonstrate Feasibility of Direct Electro-Reduction of NaBO2. (One-stepRegeneration)

Improve Sodium Electrolysis Efficiency Through Cell Design; Progressfrom Batch Synthesis to Flow Through System. (Efficiency Improvement)

Transfer Successes in Sodium Electrolysis to an Aqueous NaBO2/NaOHSystem, and Demonstrate Sodium Electrolysis Concurrent with Sodium andBoron Separation (Efficiency Improvement)

From the Different Methods, Select the Most Promising One for Economicand Engineering Study

15

Response to FreedomCAR Tech Team CommentsResponse to FreedomCAR Tech Team Comments

System storage efficiency – gravimetric and volumetric

Revised gravimetric storage efficiency by taking the average system weight ofthe fully charged state and the fully depleted state. New results included onpage 5 of this presentation.

Roadmap to improving volumetric storage efficiency pending furtherdevelopment work.

Confirm storage system cost

Initial estimates based on material cost (mostly off-the-shelf parts) forconstructing the fuel system and the first tank of fuel.

Further analysis of system cost is needed to update initial estimates.

Regeneration process energy efficiency

Experiments underway to optimize electrolysis process efficiency.