Hydrogen Production for Fuel Cells via Reformation of Coal-

Derived MethanolDr. Paul A. Erickson,

with Hyung Chul Yoon, Chang-Hsien Liao, David Davieau, J. Lars Dorr, David Vernon, Robert Kamisky, Allison Nederveld, Eddie Jordan, Michael Beerman, Zachary Zoller, Jon Hsu, Jon Otero, Todd Skinner and Matt

CaldwellUniversity of California Davis Department of Mechanical and

Aeronautical EngineeringInstitute of Transportation Studies – ITS Davis

May 2005

Project ID # PDP8This presentation does not contain any proprietary or confidential information

With Support from

2

Overview

• Project start date Oct 03• Project end date Oct 06• Percent complete 50%

• Hydrogen Production“… develop reforming

technologies for gasification and pyrolysis processes.”

-DOE technical plan

• Total project funding 500k– DOE share 400k– Contractor share 100k

• Funding received in FY04 225k• Funding received in FY05 125k

Timeline

Budget

Barriers Addressed

• Eastman Chemical• Methanex (ended

support Nov 04)

Partners

3

Motivation

Hydrogen Feedstock

Natural Gas

Methanol from Coal(Hydrogen through on-

board reforming)

Hydrogen from Coal

GasificationGasoline Eq

($/gal)3.44-4.32 1.77 3.18

Economic Study by Georgetown University and the University of Florida

For full report see http://fuelcellbus.georgetown.edu

Georgetown 1st Generation Alcohol-Fuelled Fuel Cell Bus at UC Davis

This study also serves as a baseline for bio-derived alcohol feedstocks which come from similar upstream gasification processes.

Energy Security thru a Diverse Domestic Energy Portfolio

4

Overall Objectives•Quantify the differences between coal-derived and fuel cell grade methanol (completed)•Demonstrate hydrogen production from steam reforming and autothermal reforming of coal-derived methanol (completed)•Determine hydrogen quality and conversion degradation for both coal-derived methanol and baseline fuel cell grade methanol (current)•Determine limiting steps in the reformation process when using Coal-Derived Methanol (current)•Determine and demonstrate ways to enhance the reforming methods (current)•Demonstrate and characterize operation of a hydrogen fuel cell fed by coal derived methanol (future)

5

Technical Approach• Demonstrate fuel conversion change over

time (degradation) with both coal-derived and baseline fuel

• Identify the limiting steps in the reformation processes

• Identify ways of overcoming the limiting steps in the reformation processes

• Find the relative magnitudes of each process variable on the reformation outputs including fuel type.

6

Technical Accomplishments/ Progress/Results

• Hydrogen was produced from coal-based methanol through both steam-reformation and autothermal reformation methods.

• An empirical model of steam reformer performance with coal-based methanol (Eastman) as compared to “fuel cell grade”methanol (Methanex) was developed.

• Degradation rates of reactor performance for the steam reforming method and fuels was quantified.

• Passive methods for enhancing steam reformation was investigated.

• An empirical model of the autothermal reactor performance with coal-derived methanol is being investigated.

• Transient operation of the reactors is being demonstrated.• Review of clean-up methods and capabilities of competing

methods is being analyzed for future experimental studies.

7

Milestones (on or ahead of schedule)ID Task Name Duration Start Finish

1 Peliminary Investigation And comparison of M 32.4 w Thu 1/1 Fri 8/132 External Evaluation of MeOH Fuels 4 wk Thu 1/1 Wed 1/283 Internal Evaluation of MeOH Fuels 24 wk Mon 3/1 Fri 8/134 LCMS Testing 24 wk Mon 3/1 Fri 8/135 Gravametric Testing 24 wk Mon 3/1 Fri 8/136 Analysis of Coal Based MeOH Analyte Histo 24 wk Mon 3/1 Fri 8/137 Testing 105 wkMon 12/29 Fri 12/308 Steam Reforming 105 wkMon 12/29 Fri 12/309 Setup 42 wk Mon 12/2 Fri 10/1510 Steady State 41.4 w Mon 10/18 Tue 8/211 Model Development 3 wk Mon 10/1 Fri 11/512 Data Collection 41.4 w Mon 10/1 Tue 8/213 Transient Testing 17.2 w Fri 4/1 Fri 7/2914 Degradation Testing 22 wk Mon 8/1 Fri 12/3015 Autothermal Reforming 104.2 w Thu 1/1 Thu 12/2916 Setup 48.4 w Thu 1/1 Fri 12/317 Steady State 43.2 w Mon 1/3 Mon 10/3118 Model Development 3 wk Mon 1/3 Fri 1/2119 Data Collection 39 wk Tue 2/1 Mon 10/320 Transient Testing 17.2 w Mon 5/2 Mon 8/2921 Degradation Testing 17.6 w Tue 8/30 Thu 12/222 Testing the Reformate Streams in the PEM Fue 47.6 w Mon 1/2 Wed 11/223 Preliminary Evaluation of Enhancement Requir 26 wk Mon 1/2 Fri 6/3024 PEM Fuel Cell Stack Testing 26 wk Thu 6/1 Wed 11/225 Analysis and Final Report Preparation 12 wk Mon 10/2 Fri 12/22

Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1Year 1 Year 2 Year 3 Year 4

May 2005

8

Sulfur

Chemicals

Syngas Conversion

CO2

Low Temp Gas Cooling

Shift RxCO+H2O CO2 + H2

HgRemoval(>90+%)

ParticulateScrubber

TexacoGasifier

Slag/Frit

Coal

H2O

+

ASUAir Products

O2

Slurry

Acid Gas RemovalRectisol

Sulfur RecoveryClaus/Scot

CO/H2

CO/H2Separation

H2

COFines/Char

Eastman Gasification Process

end uses of acetylLockHopper

Acetic AnhydrideAcetic AcidMethyl AcetateMethanol

From Erickson et. al. 2004 ASME Power Conference

Upstream Processes (from Eastman Chemical)

Important Background

9

R T : 0 . 0 0 - 7 0 . 0 0

0 5 1 0 1 5 2 0 2 5 3 0 3 5 4 0 4 5 5 0 5 5 6 0 6 5T i m e ( m i n )

0

5

1 0

1 5

2 0

2 5

3 0

3 5

4 0

4 5

5 0

5 5

6 0

6 5

7 0

7 5

8 0

8 5

9 0

9 5

1 0 0

Rel

ativ

e In

tens

ity

3 1 . 0 5

2 . 4 0

4 1 . 0 22 . 5 8

3 2 . 5 1 4 3 . 1 92 8 . 2 8 3 8 . 0 06 . 6 9 5 1 . 2 04 6 . 6 02 6 . 1 68 . 6 7 5 3 . 8 8 6 9 . 8 95 5 . 2 3 6 3 . 3 92 3 . 4 41 8 . 5 3

N L :6 . 4 6 E - 1A B I U V D e t e c t o r A n a l o g E r i c k s o n # 1

R T : 0 . 0 0 - 6 9 . 9 7

0 5 1 0 1 5 2 0 2 5 3 0 3 5 4 0 4 5 5 0 5 5 6 0 6 5T i m e ( m i n )

0

5

1 0

1 5

2 0

2 5

3 0

3 5

4 0

4 5

5 0

5 5

6 0

6 5

7 0

7 5

8 0

8 5

9 0

9 5

1 0 0

Rel

ativ

e In

tens

ity

3 . 0 2

3 1 . 0 7

2 6 . 1 2

3 . 7 54 1 . 1 9

2 6 . 3 7 3 4 . 6 94 . 2 3 2 3 . 5 52 1 . 4 73 7 . 1 22 7 . 5 6

3 8 . 0 74 3 . 1 31 9 . 1 6 6 9 . 7 91 6 . 6 9

N L :1 . 1 8 E - 1A B I U V D e t e c t o r A n a l o g E r i c k s o n # 4

Fuel Cell Grade

Coal-derived Methanol

Liquid Chromatography Results for both Coal-derived and Chemical Grade Methanol

Petroleum Hydrocarbons 5.9 mg/l

Petroleum Hydrocarbons 17.0 mg/l

Differences between Eastman’s Coal-derived and Fuel Cell Grade Methanol

10

Experimental Facilities

11

Steam Reforming Schematic

12

SR Reactor Geometries

Adapter for the AcousticField Generator

Schedule 40 SS Pipe,2.09 cm I.D.,(3/4” Nominal Dia.)

60.96 cm Length (24”)

Reactor A: Large Aspect Ratio (L/D=25.4) SR Reactor

To CondenserSchedule 40 SS Pipe,

2.09 cm I.D., (3/4” Nominal)12.7 cm Length

Internal Cartridge Heater 0.63 cm Dia. 20.32 cm Length

Nozzle Band Heaters

Reactor C: Cartridge Heater SR Reactor

Adapter for the Acoustic Field

Generator

Nozzle Band Heaters

Schedule 40 SS Pipe,3.51 cm I.D.(1 ¼” Nominal Dia.)

25.4 cm Length

Reactor B: Small Aspect Ratio (L/D=5.4) SR Reactor

13

The average results of multiple (three tests for each fuel) 70-hr Catalyst Degradation Tests in Reactor C for both fuel cell grade and coal-derived methanol (2.5 LHSV-M).

Degradation Tests

95

95.5

96

96.5

97

97.5

98

98.5

99

99.5

100

0 10 20 30 40 50 60 70 80

Time (Hours)

Con

vers

ion(

%)

Coal-derived Methanol

Chemical Grade Methanol

14

Hydrocarbon Concentraton

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

0 10 20 30 40 50 60 70 80

Time(hour)

Gas

Con

cent

ratio

n(%

)

Chemical Grade MethanolCoal-derived Methanol

Hydrocarbon concentrations for both fuel cell grade methanol and coal-derived methanol in multiple 70 hr degradation tests.

15

Results from two 30-hr Degradation tests in Reactor B with fuel cell grade methanol (2.5 LHSV-M). When compared to the 70-hr degradation tests in Reactor C these results show that catalyst degradation is a strong function of reactor geometry.

Chemical Grade MeOH in Reactor B

y = -0.14x + 83.786R2 = 0.9255

y = -0.1503x + 83.774R2 = 0.9701

78

79

80

81

82

83

84

0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00

Time(Hrs)

Con

vers

ion(

%)

Run 1Run 2Linear (Run 1)Linear (Run 2)

16

Understanding the Steam-Reformation

Process

Right: Typical Reactor Temperature Profile in Reactor A (Deg C). Note that the geometry is not to scale.

0.0000

0.0050

0.0100

0.0150

0.0200

0.0250

0.0300

0.0350

250 270 290 310 350

Temperature Deg C

Deg

rada

tion

(% C

onv/

hr)

7 hour performance degradation rates (% conversion / hr) in a quasi-isothermal steam reformer using fuel cell grade methanol at different temperatures. This plot shows that degradation has a strong sensitivity to temperature variations.

17

Enhancing the Steam-Reformation Process

Catalyst Bed with Two Sets of Flow Disturbers

Schematic of Fluid Pathway and Heat transfer inside a Steam Reformer Catalyst Bed

18

Temperature profiles inside the Half Radial Catalyst Bed Reactor B; Package Density of Flow Disturbers increased from Left to Right

2 PACKS 3.0 LHSV-M

4 PACKS 3.0 LHSV-M

6 PACKS 3.0 LHSV-M

8 PACKS 3.0 LHSV-M

270.00

180.00

190.00

200.00

210.00

220.00

230.00

240.00

250.00

260.00

Temp in deg C

Zone 4 Exit

Zone 3 Exit

Zone 2 Exit

Zone 1 Exit

Zone 0 Exit

0 PACKS 3.0 LHSV-M

Steam Reformation: Heat Transfer Enhancement via Flow Disturbance

19

SR Reactor PerformanceConversion vs Space Velocity (Crushed catalyst)

70

75

80

85

90

95

100

0 1 2 3 4 5

lhsv-mco

nver

sion

%

0C

2C4C

6C8C

linear (0C)

linear (2C)linear (4C)

linear (6C)linear (8C)

Conversion vs Space Velocity (Pelletized catalyst)

55

60

65

70

75

80

85

90

95

100

0 1 2 3 4 5

lhsv-m

conv

ersi

on %

0P2P4P6P8Plinear (0P)linear (2P)linear (4P)linear (6P)linear (8P)

Chemical Grade Methanol Fuel Conversion (%) versus Liquid Hourly Space Velocity of Methanol at different package density of Flow Disturbers, (1)Left: Using Pelletized

Catalyst; (2) Right: Using Crushed Catalyst

20

Autothermal Reformation

Schematic of the Autothermal-Reforming System

The Monolithic ATR Catalyst Tested

21

Autothermal Reformation of Methanol

ATR Reactor Efficiency Map from Experimental Data

•Fuel conversion was approximately 100%, when above the light off point (approx. O2/C =0.2). •Similar results are shown with coal-derived methanol.•The maximum H2 output during the experiments occurred at O2/C=0.3•The results show that the O2/C is a significant operating parameter in the ATR of methanol.•Limiting space velocity has yet to be found.•Above the light off point an equilibrium model can accurately predict the actual species concentration.•Degradation of ATR with Coal-Derived Methanol is forthcoming.

Reactor H2 Efficiency vs. GHSV0.30 O2/C

0

0.2

0.4

0.6

0.8

1

H2-E

ffici

ency

(%) .

0 10000 20000 30000 40000 50000 60000 70000

GHSV

22

Responses to Previous Year Reviewers’ Comments

• Reviewers suggested that we check the magnitude of the reactor performance degradation due to fuel impurities in relation to this same output metric due to other variables.

• We have found that reactor geometry affects the catalyst degradation in steam-reformation much more than switching from fuel cell grade methanol to coal-derived methanol. Compare degradation rates in Slide 15 (wide diameter reactor B) to the rates shown Slide 13 (Small diameter with internal cartridge heater).

23

Future Work

• Present-Oct 05– Finish degradation rate tests for fuels in Autothermal

Reactor– Finish transient tests– Review clean-up technology

• Oct 05-Oct 06 – Integrate reformer and cleanup to PEM hydrogen

fuel cell or purchase complete system– Quantify fuel cell performance with Coal-Derived vs.

Fuel Cell grade fuel

24

Conclusions and Major Findings•Coal-derived Methanol has more hydrocarbon impurities than fuel cell grade methanol. Relative levels of chlorides and sulfur are similar. (From Year 1)

•Coal-derived methanol can be used as a hydrogen feedstock with both steam reformation and autothermal reformation. Overall performance with the two fuels is comparable.

•In steam reformation with copper-based catalysts, the performance degradation with coal-derived methanol was greater than that when using fuel cell grade methanol. However, reactor geometry seems to have a much greater role in degradation than fuel impurities at this level.

•Passive flow disturbance within the steam reforming catalyst bed was investigated. From the temperature profile and fuel conversion data, it was proven that the flow disturbance made a significant heat transfer enhancement and increased the capacity of the steam reformer.

•ATR of fuel cell grade methanol has been investigated and ATR of coal-derived methanol is underway. Chemical equilibrium accurately predicts output composition above the light off point. The upper end of flow rate has not yet been determined but it is greater than 77,000 GHSV.

25

Publications and PresentationsPublished • Dorr, J. L., “Methanol Autothermal Reformation: Oxygen-to-Carbon Ratio and Reaction Progression”

Masters Thesis, UC Davis, December 2004• Dorr, J. L., and P.A. Erickson (2004), “Preliminary Modeling and Design of an Autothermal Reformer,”

Proceedings of IMECE: 2004 International Mechanical Engineering Congress and Exposition, November 13-19, 2004 Anaheim, California, IMECE 2004-59892 pp. 1-9

• Erickson, Paul A., Robert J. Kamisky and Nate Moock (2004) “Coal-Based Methanol for Use in Fuel Cells: Research Needed” proceedings of ASME POWER 2004, PWR-Vol. 35 pp. 703-710

In Press • Erickson, P.A. and H.C. Yoon (2005) “Hydrogen from Coal-Derived Methanol: Experimental Results”

Proceedings of the 3rd International Energy Engineering Conference, 2005, Paper Number AIAA-2005-5567• Liao, C.H. and P.A. Erickson, (2005)“Heat Transfer Enhancement of Steam-Reformation by Passive Flow

Disturbance Inside the Catalyst Bed” Proceedings of the ASME 2005 Heat Transfer Summer Conference San Francisco, CA, Westin St. Francis Hotel, July 17-22, 2005 Paper number HT2005-72043 Pages 1-7

In Works • C.H. Liao “An Analysis of the Effect of Flow Disturbers on Hydrogen Production via Steam-Reforming,”

Masters Thesis, UC Davis, expected June 2005.• H.C. Yoon “Hydrogen from Steam-Reformation of Coal-Derived Methanol,” Masters Thesis, UC Davis,

expected June 2005

26

The most significant hydrogen hazard associated with this project is: Build up and ignition of hydrogen gas or fuels from leaking valves or tubes

Hydrogen Safety

27

Our approach to deal with this hazard is:

Hydrogen monitoring with appropriate alarms and evacuation procedures, Automatic and Manual Safety shutoffs are included at control panel locationleak checks before and after each data run, real time monitoring and purging of hydrogen pathways before exposing personnel to the system, provide constant air flow away from reformer systems at all times, always onremoval of potential ignition sources at most likely H2 build up locations, safety training for all personnel CUPA audits maintained up to date

PI stays abreast of University, State and Federal regulations by being on Safety Committee for Mechanical and Aeronautical Engineering Department.

Hydrogen Safety