united States EPA-600/R-98-002a
Environmental Protection
AB*ncv January 1998
&EPA Research and
Development
DEMONSTRATION OF FUEL CELLS TO RECOVER
ENERGY FROM LANDFILL GAS
PHASE III. DEMONSTRATION TESTS, AND
PHASE IV. GUIDELINES AND RECOMMENDATIONS
Volume 1. Technical Report
Prepared for
Office of Research and Development
Prepared by
National Risk Management
Research Laboratory
Research Triangle Park, NC 27711
-------
FOREWORD
The U. S. Environmental Protection Agency is charged by
tecting the Nation's land, air, and water resources. Unde
environmental laws, the Agency strives to formulate and ^^^^^^m. ctcuons lead-
ing to a compatible balance between human activities and the ability of natural
systems to support and nurture life. To meet this mandate, EPA's research
program is providing data and technical support for solving environmental pro-
blems today and building a science knowledge base necessary to manage our eco-
logical resources wisely, understand how pollutants affect our health, and pre-
vent or reduce environmental risks in the future.
The National Risk Management Research Laboratory is the Agency's center for
investigation of technological and management approaches for reducing risks
from threats to human health and the environment. The focus of the Laboratory's
research program is on methods for the prevention and control of pollution to air,
land, water, and subsurface resources; protection of water quality in public water
systems; remediation of contaminated sites and groundwater; and prevention and
control of indoor air pollution. The goal of this research effort is to catalyze
development and implementation of innovative, cost-effective environmental
technologies; develop scientific and engineering information needed by EPA to
support regulatory and policy decisions; and provide technical support and infor-
mation transfer to ensure effective implementation of environmental regulations
and strategies.
This publication has been produced as part of the Laboratory's strategic long-
term research plan. It is published and made available by EPA's Office of Re-
search and Development to assist the user community and to link researchers
with their clients.
E. Timothy Oppelt, Director
National Risk Management Research Laboratory
EPA REVIEW NOTICE
This report has been peer and administratively reviewed by the U.S. Environmental
Protection Agency, and approved for publication. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Information
Service, Springfield, Virginia 22161.
-------
EFA-600/R-98-002a
January 1998
DEMONSTRATION OF FUEL CELLS TO RECOVER
ENERGY FROM LANDFILL GAS
PHASE III. DEMONSTRATION TESTS, AND PHASE IV.
GUIDELINES AND RECOMMENDATIONS
Volume 1. Technical Report
by
J. C. Trocciola
J. L. Preston
International Fuel Cells Corporation
195 Governors Highway
South Windsor, Connecticut 06074
EPA Contract 68-D1-0008
EPA Project Officer: Ronald J. Spiegel
National Risk Management Research Laboratory
Research Triangle Park, North Carolina 27711
Prepared for
U. S. Environmental Protection Agency
Office of Research and Development
Washington, B.C. 20460
-------
International Fuel Cells FCR-13524E
ABSTRACT
This report summarizes the results of a four-phase program with the U. S. Environmental Protection Agency
under Contract 68-D1 -0008, "Demonstration of Fuel Cells to Recover Energy from Landfill Gas." The envi-
ronmental impact of widespread use of this concept would be a significant reduction of global warming gas
emissions (methane and carbon-dioxide). This work was conducted over the period from January 1991
through June 1995.
International Fuel Cells Corporation (IFC) conducted the four-phase program to demonstrate that fuel cell
energy recovery using a commercial phosphoric acid fuel cell is both environmentally sound and commer-
cially feasible. Phase I, a conceptual design and evaluation study, addressed the technical and economic is-
sues associated with operation of the fuel cell energy recovery system of landfill gas. Phase II includes de-
sign, construction and testing of a landfill gas pretreatment unit (GPU) to remove critical fuel poisons such
as sulfur and halides from the landfill gas, and to design fuel cell modifications to permit operation on low
heating value landfill gas. Phase III was the demonstration test of the complete fuel cell energy recovery
system. Phase IV described how the commercial fuel cell power plant could be further modified to achieve
full rated power on low heating value landfill gas.
The demonstration test successfully demonstrated operation of the energy recovery system, including the
GPU and commercial phosphoric acid fuel cell modified for operation on landfill gas. Demonstration output
included operation up to 137kW; 37.1 percent efficiency at 120 kW; exceptionally low secondary emissions
(dry gas, 15% Oi) of 0.77 ppmV carbon monoxide, 0.12 ppniV nitrogen oxides, and undetectable sulfur diox-
ide: no forced outages with adjusted availability of 98.5 percent; and a total of 709 hours operation on landfill
gas. The pretreatment (GPU) operated for a total of 2,297 hours, including the 709 hours with the fuel cell,
and documented total sulfur and halide removal to much lower than specified <3 ppmV for the fuel cell. The
GPU flare safely disposed of the removed landfill gas contaminants by achieving destruction efficiencies
greater than 99 percent. An environmental and economic evaluation of a commercial fuel cell energy system
concluded there is a large potential market for fuel cells in this application.
-------
International Fuel Cells FCR-13524E
TABLE OF CONTENTS - VOLUME 1
Section Page
ABSTRACT ii
FIGURES vii
TABLES viii
REFERENCES ix
ABBREVIATIONS x
UNITS AND CONVERSION FACTORS x
1.0 EXECUTIVE SUMMARY 1
2.0 INTRODUCTION 7
3.0 CONCEPTUAL DESIGN, COST AND EVALUATION STUDY 9
3.1 Requirement for Landfill Gas Application 9
3.1.1 Landfill Gas Availability 9
3.1.2 Landfill Gas Characteristics 9
3.1.3 Emission Requirements 10
3.1.4 Present Options for Methane Abatement from Landfill Gas 11
3.1.5 Requirements for Conceptual Design 11
3.2 Commercial Fuel Cell Landfill Gas to Energy System Conceptual Design 12
3.2.1 Overall System Description 12
Fuel Pretreatment System 13
Fuel Cell Power Plant 16
Overall System Performance 18
Impact of Heating Value on System Performance 19
3.2.2 Environmental and Economic Assessment on the Fuel Cell Energy
Conversion System 20
Environmental Assessment 21
Economic Assessment Results 22
Comparison With Other Energy Conversion Options 24
Conclusions 25
3.2.3 Critical Issues 26
Marketing Issues 26
Technical Issues 26
-------
International Fuel Cells FCR-13524E
TABLE OF CONTENTS
Section Page
4.0 DEMONSTRATION TEST DESIGN 27
4.1 Select Landfill Site 27
4.1.1 Site Selection Criteria 27
4.1.2 Characteristics of Candidate Sites and Selection 27
4.1.3 Description of Selected Site 30
4.2 Landfill Gas Pretreatment Unit Process Design and Description 34
4.2.1 Process Operation 35
4.2.1.1 Clean Gas Production Process 35
4.2.1.2 Regeneration Process 37
4.2.1.3 Refrigeration Process 37
4.3 PC25 Power Plant Design Modifications 40
4.3.1 Introduction and Background 40
4.3.2 Phase II Summary 40
4.3.2.1 Modify Control Software 41
4.3.2.2 Cathode Exit Orifice 41
4.3.2.3 Recycle Orifice 41
4.3.2.4 Inlet Fuel Controls 41
4.3.2.5 Halide Guard Bed 41
4.3.2.6 Startup 41
4.4 Site Specific Process Design 42
4.4.1 Overall System and Site Description 42
4.5 Site Specific Engineering Design 44
4.5.1 Site Location 44
4.5.2 Site Arrangement 44
4.5.3 Site Design Details 47
5.0 GPU VERIFICATION TEST 48
5.1 Landfill Gas Pretreatment Module Test Plan 48
5.2 Permitting 50
5.2.1 South Coast Air Quality Management District Permit 50
5.2.2 L.A. City Permits 50
IV
-------
International Fuel Cells FCR-13524E
TABLE OF CONTENTS
Section Page
5.3 Test Results 51
5.3.1 Factory Test Results 51
5.3.2 Site Checkout Test Results 51
5.3.3 Phase II, EPA Field Test 52
5.3.4 Conclusions from Phase II GPU Field Test 57
6.0 FUEL CELL DEMONSTRATION TEST 58
6.1 Test and Quality Assurance Project Plan (QAPP) 58
6.2 Test Preparation 58
6.2.1 Permitting 58
6.2.2 Site Preparation 58
6.2.3 Fuel Cell Installation and Checkout on Natural Gas 58
6.2.4 Modifications for Landfill Gas 60
6.2.5 Checkout for Landfill Gas Operation 60
6.3 Demonstration Test Results 61
6.3.1 GPU Performance 61
6.3.1.1 Operation and Reliability 61
6.3.1.2 GPU Contaminant Removal Performance 63
6.3.1.3 GPU Exit Gas Heat Content 65
6.3.2 Fuel Cell Performance 66
6.3.2.1 Fuel Cell Operation and Availability 66
6.3.2.2 Fuel Cell Power Plant Efficiency 67
6.3.2.3 Fuel Cell Maintenance and Operator Requirements 69
6.3.3 Emissions 70
6.3.4 Quality Assurance 71
7.0 PHASE IV GUIDELINES AND RECOMMENDATIONS 74
8.0 CONCLUSIONS 76
-------
Internationa! Fuel Cells FCR-13524E
TABLE OF CONTENTS - VOLUME 2
Section Page
LIST OF APPENDICES
Appendix A (Summary of Detailed Site Design) A-l
Appendix B (Landfill Gas Pretreatment Test Plan) B-l
Appendix C (H2$ Removal Over Westates Carbon) C-l
Appendix D (Executive Summary of Landfill Gas Pretreatment Performance Test
Report by TRC Environmental Corp.) D-1
Appendix E (Properties of d-limonene Refrigerant) E-l
Appendix F (Laboratory Data on Reaction of Hydrogen Sulfide to Carbonyl Sulfide) F-l
Appendix G (Site Specific Test Plan and Quality Assurance Project Plan, Revision
No. 2, December 1994) G-l
Appendix H (System Performance and Emission Test Report, by TRC Environmental,
May 1995) Phase III Fuel Cell/Landfill Gas Energy Recovery Demonstration,
Penrose Landfill H-l
Sub-Appendix A - Process Data H-A1
Sub-Appendix B - GPU Exit Heat Content Analytical Data - ASTM Method H-B1
Sub-Appendix C - Power Plant Emissions Data H-C1
Sub-Appendix D - Flare Emission Data From Phase II H-D1
Sub-Appendix E - GPU Exit Contaminant Measurement Data H-E1
Sub-Appendix F- Calibration Data And Certifications H-F1
Sub-Appendix G - ASTM Method Heat Content Analysis QA Replicates H-G1
Sub-Appendix H - Halide And Sulfur Compound Audit Data H-H1
Sub-Appendix I - Fuel Cell Emissions QA Data H-I1
Sub-Appendix J - Fuel Cell Emissions Calibration Error Data H-J1
Sub-Appendix K - Fuel Cell Exhaust Gas Flowrate Data H-K1
Sub-Appendix L - ASTM Heat Content Analysis Audit Data H-L1
VI
-------
International Fuel Cells FCR-13524E
LIST OF FIGURES
Figure Page
1-1. Fuel Cell Energy Conversion System Commercial Concept 2
1-2. Landfill Gas Pretreatment Unit (GPU) System 3
1-3. GPU Installation at Pacific Energy Landfill 4
1-4. PC25 Power Plant Installation at California Landfill Site 6
3-1 Commercial Fuel Cell Landfill Gas to Energy Conversion Concept 12
3-2. Simplified Block Diagram of Commercial LFG Pretreatment System 13
3-3. Staged Regeneration of Adsorbent Beds and Sample Regeneration Sequence 15
3-4. Functional Schematic Fuel Cell Landfill Gas Power Unit 17
3.5. Overall System Schematic and Performance Estimate for Fuel Cell
LFG to Energy Conversion System 19
3-6. Impact of Landfill Gas Heating Value on Power Plant Power Output
and Heat Rate 20
3-7. Comparison of Fuel Cell to Flare for Methane Mitigation
Assuming Electric Revenues, Emission Credits and Thermal Recovery 23
3-8. Comparison of Fuel Cell to Flare for Methane Mitigation Assuming Electric
Revenues and Emission Credits 23
3-9. Comparison of Fuel Cell to Flare for Methane Mitigation Assuming Electric
Revenues Only 24
3-10. Comparison of Fuel Cell to I.C.E. Energy Conversion System 25
4-1. Penrose Plant Supplies Alternative Energy to Southern California Power
Grid (Courtesy of Pacific Energy) 31
4-2. Landfill Gas to Electric Power (Courtesy of Pacific Energy) 32
4-3. Fuel Cell Site Options (Courtesy of Pacific Energy) 33
4-4. Landfill Gas Pretreatment Unit System 35
4-5. Gas Purification Process 36
4-6. Regeneration Process 38
4-7. Refrigeration Process Unit 39
4-8. LFG Fuel Cell Demonstration Program 42
4-9. Demonstration Project Processes 43
4-10. Fuel Cell Site Options: Site 2 Selected for Demonstration
(Courtesy of Pacific Energy) 45
4-11. Site Layout 46
5-1. Phase II Gas Pretreatment Unit Sample Location 53
5-2. Landfill Gas Pretreatment Unit Sample Location for GPU Flare Tests 55
6-1. Installation of PC25 at Los Angeles Landfill 59
6-2. Photograph of the GPU and Power Plant Installed at the Penrose Site 60
6-3. GPU Exit Contaminant Concentration vs. Time 63
6-4. Demonstrator System Schematic 68
7-1. PC25 C Fuel Delivery Train 74
7-2. PC25 C Fuel Delivery Train Modified for Operation on Landfill Gas 74
vu
-------
International Fuel Cells FCR-13524E
LIST OF TABLES
Table Page
1-1. GPU Sulfur and Halide Contaminant Removal Performance and Specification 5
3-1. Size Distribution of Landfills and Potential Power Output 9
3-2. Landfill Gas Characteristics 10
3-3. Key Features of Commercial Pretreatment System Conceptual Design 15
3-4. Gas Pretreatment System Projected Performance 16
3-5. Performance Comparison for Nominal 200 kW Output 17
3-6. Estimated Fuel Cell Air Emissions 18
3-7. Site Characteristics for Landfill Gas Assessment 21
3-8. Emissions Impact of Fuel Cell Energy Recovery from Landfill Gas 21
4-1 Pacific Energy Landfill Gas Sites 28
4-2 Assessment of Candidates Sites vs. Evaluation Criteria 29
4-3 Supplemental Landfill Data for Candidate Sites 30
4-4 Raw Landfill Gas Contaminants and Concentration at Penrose Test Site 34
4-5 Modification to PC25 A for Operation at 140 kW in Landfill
Gas Demonstration 40
4-6 Summary of Detail Site for EPA Landfill Gas Demonstration 47
5-1 Test Protocol for Phase II EPA Field Test 49
5-2 Permit Activities for EPA Gas Pretreatment 50
5-3 Gas Pretreatment Unit Sulfur Removal Performance 52
5-4 Summary of Phase II Testing of Gas Pretreatment Unit 54
5-5 Landfill Gas Pretreatment Unit Field Test Results for GPU Flare 55
6-1 GPU Validation Test Results Prior to Start of Fuel Demonstration Field Test 61
6-2 GPU Run Summary 62
6-3 GPU Contaminant Removal Performance During Phase III 64
6-4 GPU Exit Gas Heat Content 65
6-5 Summary of Fuel Cell Operations on Landfill Gas 66
6-6 Fuel Cell Electrical Efficiency on Landfill Gas 68
6-7 Operation and Maintenance Cost Factor for Commercial Applications 69
6-8 Fuel Cell Emissions Summary on Landfill Gas 70
6-9 Summary of Quality Assurance Goals and Test Results 72
6-10 Typical Concentrations, Detection Limits, and Blank Results for Targeted
Compounds in the Raw Landfill Gas at the Penrose Landfill 73
vin
-------
International Fuel Cells FCR-13524E
REFERENCES
1. Air Emissions from Municipal Solid Waste Landfills - Background Information for Proposed Standards
and Guidelines, EPA-450/3-90-011a (NTIS PB91-197061). March 1991, page 3-30.
2. Landfill Gas Utilization - Database of North American Projects, Susan A. Thomloe and John G. Pacey,
presented at the Solid Waste Association of North America's 17th Annual International Landfill Gas
Symposium, March 22-24, 1994, Long beach, CA.
3. Demonstration of Fuel Cells to Recover Energy from Landfill Gas, Phase I Final Report: Conceptual
Study (Report EPA-600/R-92-007; NTIS PB92-137520). G. J. Sandelli, January 1992.
4. Solid Waste & Power, "Will Gas-To-Energy Work at Your Landfill?," Greg Maxwell, June 1990, p.44.
5. Air Emissions for Municipal Solid Waste Landfills - Background Information for Proposed Standards
and Guidelines, EPA-450/3-90-01 la (NTIS PB91-197061). March 1991, page 3-23.
6. Ibid, Table 3-6 pages 3-25 through 3-28.
7. "Recovery of VOC's Using Activated Carbon;" James R. Graham, and Mukuno Ramaratnam; Chemical
Engineering, February 1993.
IX
-------
Internationa] Fuel Cells
FCR-13524E
ABBREVIATIONS
EPA
IFC
ONSI
MSW
NMOC
SCR
GPU
QAPP
SCAQMD
LADWP
LFG
United States Environmental Protection Agency
International Fuel Cells Corporation
A Subsidiary of IFC (from On-Site Power)
Municipal Solid Waste
Non Methane Organic Compound
Selective Catalytic Reduction
Gas Pretreatment Unit
Quality Assurance Project Plan
South Coast Air Quality Management District
Los Angeles Department of Water and Power
Landfill Gas
UMTS AND CONVERSION FACTORS
POWER
MW
kW
MASS
Mg
Tg
VOLUME
SCMD
SL'M
PRESSURE
Pa
HEATING VALUE
kcal/SL
Megawatt
Kilowatt
Megagrams (106 grams)
Terragrams (109 grams)
Standard cubic meters per day
Standard liter per minute
Pascal
Kilocalories per standard liter
To Convert To
pounds
pounds
SCFD (std cubic feet/
day)
SCFM (std cubic feet/
min)
PSI
Btu/SCF
Multiply
By
2,205
2,204,600
35.3
0.0353
1.45 x 10-4
112
-------
International Fuel Cells FCR-13524E
1.0 EXECUTIVE SUMMARY
The US Environmental Protection Agency (EPA) has promulgated standards and guidelines for the control
of air emissions from municipal solid waste (MSW) landfills. This Clean Air Act regulation will result in
the control of up to 7 Tg/year of CH4. The collection and disposal of waste methane, a significant contributor
to the greenhouse effect, would result from the emission regulations. This EPA action provides an opportuni-
ty for energy recovery from the waste methane that could further benefit the environment. Energy produced
from landfill gas could offset the use of foreign oil, and air emissions affecting global warming, acid rain,
and other health and environmental issues.
International Fuel Cells Corporation (IFC) was awarded a contract by the US EPA to demonstrate energy
recovery from landfill gas using a commercial phosphoric acid fuel cell. IFC conducted a three-phase pro-
gram to show that fuel cell energy recovery is environmentally feasible in commercial operation. Work was
initiated in January 1991. Phase I, a conceptual design and evaluation study, addressed the problems
associated with landfill gas as the feedstock for fuel cell operation.
Phase II of the program included construction and testing of the landfill gas pretreatment module to be used
in the demonstration. Its objective was to determine the effectiveness of the pretreatment system design to
remove critical fuel cell catalyst poisons such as sulfur and halides.
Phase III of this program was a demonstration of the complete fuel cell energy recovery concept.
Phase IV prepared guidelines and recommendations describing how the PC25 ~ C power plant could be mo-
dified to achieve full-rated power of 200 kW on landfill gas, based upon the experience gained testing the
PC25 A Model in this program.
Phase I
The MSW landfills in the US were evaluated to determine the potential power output which could be derived
using a commercial 200 kW fuel cell. Each fuel cell would consume 2800 SCMD of landfill gas to generate
200 kW, assuming a heating value of 4.45 kcal/liter.
The potential power generation market available for fuel cell energy recovery was evaluated using an EPA
estimate of methane emissions in the year 1992 l. An estimated 4370 MW of power could be generated from
the 7480 existing and closed sites identified. The largest number of potential sites greater than 200 kW occurs
in the 400 to 1000 kW range. This segment represents a market of 1700 sites or 1010 MW.
The Phase I assessment concluded that these sites are ideally suited to the fuel cell concept. The concept can
provide a generating capacity tailored to the site because of the modular nature of the commercial fuel cell.
The best competing options, Rankine and Brayton Cycles, are not as effective at these power ratings due to
high emission and poor energy utilization..
As a result of the assessment, the conceptual design of the commercial concept was required to be modular
(transportable from site-to-site) and sized to have the broadest impact on the market. The design is based
on providing a modular, packaged, energy conversion system which can operate on landfill gases with a wide
range of compositions as typically found the US. The complete system incorporates the landfill gas collection
system, a fuel gas pretreatment system and a fuel cell energy conversion system. In the fuel gas pretreatment
section, the raw landfill gas is treated to remove contaminants to a level suitable for the fuel cell energy con-
version system. The fuel cell energy conversion system converts the treated gas to electricity and useful heat.
1. Air Emissions from Municipal Solid Waste Landfills - Background Information for Proposed Standards and
Guidelines, EPA-450/3-90-01 la (NTIS PB91-197061). March 1991, page 3-30.
-1-
-------
International Fuel Cells
FCR-13524E
Landfill gas (LFG) is utilized in 110 MSW landfills in the US2. These systems have proven the effectiveness
of the landfill gas collection systems. Therefore design and evaluation studies in Phase I were focused on
the energy conversion concept utilizing fuel cells.
The commercial landfill gas to energy conversion system is illustrated in Figure 1-1. The fuel pretrealment
system has provisions for handling a wide range of gas contaminants. Multiple pretreatment modules can
be used to accommodate a wide range of landfill sizes. The wells and collection system collect the raw landfill
gas and deliver it at approximately ambient pressure to the gas pretreatment system. In the gas pretreatment
system the gas is treated to remove NMOCs including trace constituents which contain halogen and sulfur
compounds.
Landfill gas wells
and collection
system
800 kW Fuel cell power plant
operating on landfill gas
Utility
grid
\
Landfill site
office and
blower
Gas
pretreatment
system
Multiple
fuel cell
power plants
FC30744 0
H94100B
Figure 1-1. Fuel Cell Energy Conversion System Commercial Concept
The commercial energy conversion system shown in Figure 1-1 consists of four fuel cell power plants. These
power plants are designed to provide 200 kW output when operating on landfill gas with a heating value of
4.45 kcal/liter and for accommodating higher contaminant concentrations. The output from the fuel cell is
utility grade ac electric power. It can be transformed and put into the electric grid, used directly at nearby
facilities, or used at the landfill itself. The power plants are capable of recovering co-generation heat for
nearby use or rejecting it to air.
2. Landfill Gas Utilization - Database of North American Projects, Susan A. Thornloe and John G. Pacey,
presented at the Solid Waste Association of North America's 17th Annual International Landfill Gas Sympo-
sium. March 22-24. 1994, Long beach, CA.
-2-
-------
International Fuel Cells
FCR-13524E
Phase II
The major element of Phase II was the construction and subsequent testing of a gas cleanup system at the
Penrose Landfill site in Los Angeles (Sun Valley), California. Landfill gases consist primarily of carbon
dioxide (CO2), methane (OHU), and nitrogen (Nz), plus trace amounts of hydrogen sulfide (H2S), organic
sulfur, organic halides and non-methane hydrocarbons. The specific contaminants in the landfill gas of con-
cern to the fuel cell are sulfur and halides. Both of these ingredients can "poison" and therefore reduce the
life of the fuel cell power plant's fuel processor. The fuel processor converts CHU in the landfill gas stream
into hydrogen (H2> and CO2 in an endothermic reaction over a catalyst bed. The catalyst in this bed can react
with the halides and sulfides and lose its activity; i.e., poison irreversibly.
The system designed to remove fuel cell contaminants is shown in Figure 1-2. This system is known as the
Gas Pretreatment Unit (GPU). H2S is first removed by adsorption on a packed bed. The material which per-
forms this function is a specially treated carbon activated to catalyze the conversion of P^S into elemental
sulfur which is deposited on the bed. This conversion to sulfur is by the following reaction:
H2S 4 1/2 O2
H2O
This bed is not regenerable on site, but the carbon can be regenerated off site if desired.
LFG ^
HjS
AOSOR8EH
mi* (*
S
COOLER
CONDENSER
1
1
~^
/
CONDENSATION
OF WATER AND
ORVER
=>
LOW
TEMPERATURE
COOLER
ADSORPTION
OF WATER
HYDROCAR-
BOO
IS
^>
ACTIVATED
CARBON
' j PARTICULATE "
7 FILTER
ADSORPTION OF
HYDROCARBONS INCLUDING ORGANIC
SLILFUO AND HALOGEN COMPOUNDS
260'C
REGENERATION
TO .
FLARE
WATER
DESORPTION
TO ป
FLARE
260"C
REGENERATION
H/C
DESOflPTION
^> CLEAN
r-TT LFG
TO
FUEL
CELL
REGENERATION
II.BhlerKsec
I
HP295-01
R951104
Figure 1-2. Landfill Gas Pretreatment Unit (GPU) System
The first stage cooler removes water, some heavy hydrocarbons, and sulfides which are discharged as conden-
sate to the Penrose plant's existing water condensate pretreatment system. Since the demonstration landfill
GPU operates on a small slipstream from the Penrose site compressor and gas cooler, some of the water and
heavy hydrocarbon species are removed prior to the GPU. Most of the contaminant halogen and sulfur spe-
cies are lighter and remain in the landfill gas to be treated in the gas pretreatment unit. All remaining water
in the landfill gas, as well as some sulfur and halogen compounds, are removed in a regenerable dryer bed
which has a high capacity for adsorbing the remaining water vapor in the landfill gas. There are two dryer
beds so that one is always operational while the other is being regenerated. The dry landfill gas is then fed
to the second stage cooler. This cooler can be operated as low as -32ฐ C and potentially can condense out
-3-
-------
International Fuel Cells
FCR-13524E
additional hydrocarbons if present at high enough concentrations. In addition, the second stage cooler re-
duces the temperature of the carbon bed, therefore enhancing its adsorption performance. The downstream
hydrocarbon adsorption unit, whose temperature is controlled by the second stage cooler, is conservatively
sized to remove all heavy hydrocarbon , sulfur and halogen contaminant species in the landfill gas. This unit
consists of two beds of activated carbon so that one is always operational while the other is being regenerated.
Both the regenerable dryer and hydrocarbon removal beds operate on a nominal 16 hour cycle of each set of
beds operating in the adsorption mode for 8 hours and regeneration mode for 8 hours. The gas then passes
through a paniculate filter and is warmed indirectly by an ambient air finned tube heat exchanger to ensure
a fuel inlet temperature above 0ฐ C before being fed to the fuel cell unit.
The GPU was constructed by IFC at its facility in South Windsor, Connecticut. Construction of the unit was
completed in February 1993. Upon completion of construction, the unit was evaluated at the South Windsor
facility, using N2 as the test gas. The unit successfully completed the 16 hour control test verifying that rated
flows, pressure, and temperature were achieved. After the test, the unit was shipped to the landfill site located
in Los Angeles, California, where it was installed in April 1993. Figure 1-3 is a photograph of the unit
installed at the site.
Figure 1-3. GPU Installation at Pacific Energy Landfill
The GPU was successfully tested at the Penrose landfill site in Los Angeles (Sun Valley), California. The
GPU successfully removed the sulfur and halogen compounds contained in the landfill gas to a level signifi-
cantly below the specified value for use with the phosphoric acid fuel cell and to date has operated for approxi-
mately 2300 hours.
Table 1 -1 compares the measured sulfur and halide contents of the gas produced by the GPU to the specifica-
tion value. The data verify that the GPU reduces the sulfur and halide contents of landfill gas to a concentra-
tion lower than required by the fuel cell power plant. The exceptionally low GPU exit contaminant levels
indicate that the low temperature cooler is not essential, even thought the reduced temperature in the activated
carbon bed increases capacity for sulfur and halogen compounds. For system simplification in the future,
it may be beneficial to eliminate the low temperature cooler, and simplify the refrigeration system, in ex-
-4-
-------
International Fuel Cells
FCR-13524E
change for increasing the activated carbon bed volume slightly. Based on the favorable results of the GPU
testing, the EPA directed IFC to proceed into Phase III of the program, which entails characterizing the perfor-
mance; i.e., emissions, efficiency, and power output of the commercial phosphoric acid fuel cell power plant
when operating on land fill gas which has been purified by the GPU.
Table 1-1.
GPU Sulfur and Halide Contaminant Removal Performance and Specification (ppmV)
Total Sulfur (as H2S)1
in ppmV
Total Hal ides (as Chloride)2
in ppmV
INLET
117
47
OUTLET
<0.047
<0.032
SPECIFICATION
<3
<3
1 Measured by Gas Chromatographyt 'Flame Photometric Delineation by EPA Methods 15, 16. and 18
2 Measured by Gas Chromatographyl by EPA Method TO-14
The power plant utilized in this program is a commercial PC25 ~ 200 kW phosphoric acid fuel cell. The pow-
er plant was shipped and installed at the Penrose Landfill during 1994 (Figure 1 -4). The unit was started on
natural gas prior to its modification for operation on landfill gas. This testing was conducted to establish a
baseline performance level. Upon completion of the natural gas testing, the unit was shut down, modified
for low heating value gas, and subsequently connected to the GPU for testing on landfill gas. All power pro-
duced by the unit was fed into the electrical grid for sale to the local electrical utility, the Los Angeles Depart-
ment of Water and Power (LADWP). This fuel cell is the first ever connected to the LADWP utility system
grid. The revenue produced by the sale of this electricity was used to help offset program costs.
Emission testing of the power plant effluent was conducted by TRC Environmental Corporation during Feb-
ruary 1995. Using EPA Methods 6c, 7e and 10 respectively, emission levels of sulfur dioxide (SOz) were
undetectable at a detection limit of 0.23 ppm, while nitrogen oxides (NOX) averaged 0.12 ppm and carbon
monoxide (CO) averaged 0.77 ppm. All the data are dry measurements corrected to 15% oxygen (02). These
emission levels verify that fuel cells can operate on landfill gas while maintaining the low emission levels
characteristic of this commercial fuel cell power plant.
An exciting dimension of the PC25 operating on landfill gas is that, unlike internal combustion engines and
turbines, the unit has significant siting characteristics due to its demonstrated low levels of emissions, noise
and vibration. It can be located remote from the landfill using gas piped from the site. In this way, its thermal
energy, as well as its power, can be put to constructive use at a customer's building. In addition, by siting
at the building, the economics improve significantly since the power plant displaces commercial electricity
which has a much higher cost than the revenue which would be received if the fuel cell were sited at a landfill
and received utilities' "avoided" cost. Utilizing the fuel cell's thermal energy can result in an overall efficien-
cy [i.e., (Electrical Energy plus Thermal Energy)/Energy Content of Gas Consumed] of 80%. This high effi-
ciency conserves natural resources and reduces the amount of CO2 emitted to the atmosphere. It also im-
proves the economics, since heat may be sold to the building owner.
-5-
-------
International Fuel Cells
FCR-13524E
WCN-15074
Figure 1-4. PC25 Power Plant Installation at California Landfill Site
-6-
-------
International Fuel Cells FCR-13524E
2.0 INTRODUCTION
This report summarizes the results of a four-phase program with the U.S. Environmental Protection Agency
under Contract 68-DI-0008, "Demonstration of Fuel Cells to Recover Energy from Landfill Gas." The envi-
ronmental impact of widespread use of this concept would be a significant reduction in global warming gas
emissions (methane and carbon dioxide). This work was conducted over the period from January 1991
through June 1995.
The results of the Phase I activity from January 1991 to August of 1992 are summarized in Section 3 of this
report. In Phase I, a conceptual design of a commercial scale landfill-gas-to-energy concept utilizing a com-
mercial phosphoric acid fuel cell is established.3 This conceptual design is utilized to identify key issues
associated with utilizing landfill gas as a feedstock for fuel cell operation, and to establish a conceptual design
for the landfill-gas-to-energy demonstration utilizing a phosphoric acid fuel cell, conducted in Phase III of
the program.
The Phase II activity for the period September 1992 through December 1993 is discussed in Section 4 and
5. The objective in Phase II is to address the two major technical issues impeding commercialization of the
commercial concept: 1) Cleanup of the landfill gas to a level suitable for the fuel cell power plant; 2) Modifi-
cation of the fuel cell power plant for operation on the dilute methane landfill gas fuel. These issues are ad-
dressed in a detail site specific process and engineering design of the system design, described in Section 4
of this report, and the construction and test of the gas pretreatment system which is described in Section 5
of this report.
Section 4 of the report, titled Demonstration Test Design, describes the site specific process and engineering
design of the gas pretreatment unit designed for this application, plus the PC25 A, 200-kW phosphoric acid
fuel cell power plant, and the landfill-gas-to-energy site which was selected in Phase I. The PC25 A fuel cell
is uniquely suited to this application due to its low secondary emissions: typically 0.5 ppmV NOX, 1.1 ppmV
carbon monoxide; and 0.03 ppmV non-methane hydrocarbons (all measured at 15 percent oxygen on a dry
gas basis using natural gas fuels). Section 4.1 describes the development of the gas pretreatment process from
a conceptual design in Phase I, to a complete, detailed mechanical and process design including field modifi-
cations in Phase II. In Section 4.2 the PC25 A power plant design modifications to permit operation on land-
fill gas up to a nominal 140 kW rating are described. Section 4.3 describes the overall site specific process
design for the demonstration including the gas pretreatment system and its integration to the fuel cell power
plant system. Section 4.4, Site Specific Engineering Design, summarizes the details of the site location and
construction for the demonstration equipment.
Section 5, Gas Pretreatment Unit Verification Test, summarizes all aspects of the verification testing of the
gas pretreatment system during Phase II. Section 5.1 summarizes the Landfill Gas Pretreatment Module Test
Plan and the test protocol used to direct all test activities, while Section 5.2 reviews permitting requirements
including South Coast Air Quality Management District and City of Los Angeles Building and Safety De-
partment. Section 5.3 reviews test result including factory testing, initial field checkout testing, and the field
verification test.
The results of the Phase III demonstration test activities beginning on January 1994 and ending in June, 1995
are described in Section 6. During this third phase of the program, IFC developed a Test and Quality Assur-
ance Project Plan, completed all permitting activities for the fuel cell with the City of Los Angeles, installed
and checked out the fuel cell power plant at the site on natural gas, modified the fuel cell for operation on
landfill gas, and then connected the fuel cell to the gas pretreatment unit and operated the demonstration test,
including obtaining critical emissions and operating data. The demonstration operated at the existing Penrose
Station landfill gas energy recovery facility owned by Pacific Energy in Sun Valley, California. Internal com-
bustion engines presently generate up to 8.9 MW of electricity at this site with landfill gas from four separate
3. Demonstration of Fuel Cells to Recover Energy from Landfill Gas, Phase I Final Report: Conceptual
Study (Report EPA-600/R-92-007; NTIS PB92-137520). G.J. Sandelli, January 1992.
-7-
-------
Internationa] Fuel Cells FCR-13524E
landfills. Electricity produced by the fuel cell was sold to the Los Angeles Department of Water and Power
electric utility grid.
The QAPP is described in Section 6.1 of this report. The field test preparation is described in Section 6.2,
including permitting, site preparation, fuel cell installation and checkout on natural gas, fuel cell modification
for landfill gas, and checkout of the GPU plus fuel cell operating together on landfill gas. Section 6.3 summa-
rizes the field test results, including GPU and fuel cell performance, and fuel cell emissions.
The results of the Phase IV study to identify the lowest cost means to modify the latest model ONSI commer-
cial phosphoric acid fuel cell, the PC25 C, for operation at a full rated 200 kW using landfill gas are summa-
rized in Section 7.
The conclusions for the four-phase program are given in Section 8.
-8-
-------
International Fuel Cells
FCR-13524E
3.0 CONCEPTUAL DESIGN, COST AND EVALUATION STUDY RESULTS
3.1 Requirement for Landfill Gas Application
This section reviews the opportunities for using fuel cells for methane mitigation and energy conversion and
describes the significant potential market for power generation using fuel cells. A list of requirements is de-
veloped for the conceptual design of a commercial fuel cell landfill gas to energy system. The results of the
evaluation study form the basis for a conceptual design of a demonstrator fuel cell system for testing at a se-
lected landfill gas site.
3.1.1 Landfill Gas Availability
The Municipal Solid Waste (MSW) landfills in the United States were evaluated to determine the potential
power output which could be derived using a commercial 200-kW fuel cell. Each fuel cell would consume
2800 SCMD of landfill gas to generate 200 kW, assuming a heating value of 4.45 kcal/liter.
The potential power generation market available for fuel cell energy recovery was evaluated using an EPA
estimate of methane emissions in the year 19971 and an estimate of landfill gas production rate of 3.08 liters
per Mg per year of refuse in place4. An estimated 4370 MW of power could be generated from the 7480 exist-
ing and closed sites identified as shown in Table 3-1. The largest number of potential sites greater than 200
kW occurs in the 400 to 1000 kW range. This segment represents a market of 1700 sites or 1010 MW.
The assessment concluded that these sites are ideally suited to the fuel cell concept. The concept can provide
a generating capacity tailored to the site because of the modular nature of the commercial fuel cell. The best
competing options, Rankine and Brayton Cycles are not as effective at these power ratings due to high emis-
sions and poor energy utilizations.
The result of our assessment is a requirement for the conceptual design of the commercial concept to be modu-
lar in nature and sized to have the broadest impact on the market.
Table 3-1. Size Distribution of Land fills
and Potential Power Output
Site Power Rating (kW)
Less than 200
201-400
401-1000
1001-1500
1501-2000
2001-2500
2501-3000
>3000
Total
No. of Sites
3700
1100
1700
380
220
90
60
230
7480 Sites
Total Power
Output (MW)
220
330
1010
480
380
190
160
1600
4370 MW
3.1.2 Landfill Gas Characteristics
The available information on landfill gas compositions was evaluated to determine the range of gas character-
istics which a fuel cell landfill-gas-to-energy power plant will encounter. This information was used to set
the requirements for the gas pretreatment and fuel cell power plant designed to operate on a wide range of
available landfill gas compositions within the United States.
4. Solid Waste & Power. "Will Gas-To-Energy Work at Your Landfill?," Greg Maxwell, June 1990, p.44.
-9-
-------
International Fuel Cells
FCR-13524E
A summary of landfill gas characteristics is shown in Table 3-2. The heating value of the landfill gas varies
from 3.12 to 5.34 kcal per liter with a typical value of 4.45 kcal per liter. The major non-methane constituent
of landfill gas is carbon dioxide. The carbon dioxide ranges from 40 to 55 percent of the gas composition
on a dry basis. Other diluent gases include nitrogen and oxygen, which are indicative of air incursion into
the well (most frequently in perimeter wells). Nitrogen concentrations can range as high as 15 percent on
a dry basis but typical values are five percent or less. Oxygen concentrations are monitored closely and held
low for safety reasons. Pacific Energy has indicated that the landfill gas is typically saturated with water va-
por at temperatures up to 49 ฐC.
Table 3-2. Landfill Gas Characteristics
Characteristic
Heating Value
(HHV)
CH4
CO2
N2
O2
Sulfur as H2S
Halides
Non-Methane Organic Com-
pounds (NMOCs)
Range
3.12-5.34
kcal/1
35-58%
40-55%
0-15%W
0-2.5%(D
1-700 ppmv
N/A
237-14,294 ppmv
(as hexane)
Typical
4.45
kcal/1
50%
45%
5%
<1% (for safety)
21 pprnv
132 ppmv
2700 ppmv
(as hexane)
Note: ( I ) Highest values occur in perimeter wells
Landfill gas contains trace amounts of nonmethane organic compounds (NMOCs). A typical value of NMOC
concentration of 2700 ppmv (expressed as hexane) was derived from data provided by EPA5. The NMOC
concentration in the landfill gas is an important measure of the total capacity required in the gas pretreatment
system, while the specific individual analyses provide a basis for gas pretreatment subcomponent sizing. The
specific contaminants in the landfill gas, of interest to the fuel cell, are sulfur and halides (chiefly chlorides
and fluorides). The sulfur level ranges from 1 to 700 ppmv, with a typical value on the order of 21 ppmv.
Sufficient data were not available to assess the range of the halides, but a typical value of 132 ppmv was calcu-
lated for this contaminant6. The range of contaminant values varies not only from site to site, but also at any
given site with time due to seasonal weather or moisture content. These characteristics require the pretreat-
ment system design to be capable of handling these gas quality variations to avoid expensive site specific
engineering of the pretreatment design which would affect the marketability and economics of the concept.
3.1.3 Emission Requirements
Existing U.S. regulations do not address methane emissions from landfills directly. Proposed new EPA regu-
lations would control non-methane organic compounds from large landfills (150 Mg per year and up) and
hence would indirectly control methane emissions.
Landfill gas emission requirements are primarily determined at the state and local level. State requirements
are generally limited to controlling explosion hazards, typically limiting methane concentrations to below
25 percent of the lower explosion limit. An evaluation of state regulations revealed that collection and control
requirements generally necessitate venting, or the use of a flare. However, Federal Clean Air Act require-
ments are driving the state and local air quality rules toward tighter controls, including secondary air emis-
5. Air Emissions for Municipal Solid Waste Landfills - Background Information for Proposed Standards and
Guidelines, EPA-450/3-90-011a (NTIS PB91-197061). March 1991, page 3-23.
6. Ibid, Table 3-6 pages 3-25 through 3-28.
-10-
-------
International Fuel Cells FCR-13524E
sions which would result from energy recovery processes. For instance, in non-attainment regions for ozone,
strict requirements for secondary emissions including NOX, carbon monoxide, and NMOCs may exist. The
best known example of such strict local emission requirements is the South Coast Air Quality Management
District (SCAQMD) in southern California.
3.1.4 Present Options for Methane Abatement from Landfill Gas
A number of landfill gas methane abatement options exist, each with its own particular characteristics and
range of applicable economic landfill sizes or site characteristic for optimum use. Among these options the
fuel cell is unique in that it produces electric power and recovers waste heat at higher efficiency and produces
negligible secondary emissions. The fuel cell performance allows it to be used efficiently in small sites, while
its modularity allows its use in larger sites covering a significant portion of the landfill gas market. The char-
acteristic of modularity allows the fuel cell to match the landfill gas output of the site as production expands
or is depleted. A review of the abatement options indicates that the fuel cell should be evaluated and com-
pared competitively for small (< 1 MW) to medium (< 3 MW) capacity sites with the most common means
of mitigating methane, the flare, and the lean-burn internal combustion engine. The internal combustion en-
gine can be modified for cogeneration and/or secondary emission reduction with the addition of selective
catalytic reduction (SCR) with ammonia and gas pretreatment to protect the emission catalyst from the con-
taminants in landfill gas. The fuel cell system was not compared to turbine technologies in this study because
the characteristics of the fuel cell system are attractive to smaller sites and thus the economics cannot be fairly
compared. Combustion turbines, however, are an effective abatement option for larger capacity sites (> 3
MW).
3.1.5 Requirements for Conceptual Design
A competitive fuel cell system for abating landfill gas methane can provide an attractive, low emission, flex-
ible and cost effective alternative-to present mitigation and energy conversion systems.
The conceptual design of a landfill gas fuel cell conversion system must incorporate those features which can
provide this capability and be verified in a demonstration program. To meet this potential the conceptual
design must incorporate those features which meet the following requirements:
Application to a Large Number of Landfills - The conceptual design can accommodate a wide
range of landfill sizes through the use of a basic building block or modularity and multiples of
these modules.
Accommodation to Variations in Landfill Gas Composition and Contaminant Level - The
landfill gas pretreatment and fuel cell system is tolerant to variations in gas heating value and
a wide range of contaminant compositions.
Competitive Economics - Cost to mitigate methane and NMOC emissions from MSW landfills
to proposed EPA regulations should be minimized.
Low Emissions - The overall system air emissions, solid and liquid wastes are kept at a mini-
mum.
-11-
-------
International Fuel Cells
FCR-13524E
3.2 Commercial Fuel Cell Landfill Gas to Energy System Conceptual Design
This section describes the commercial fuel cell landfill gas to energy system conceptual design. The concep-
tual design is based on providing a modular, packaged, energy conversion system which can operate on land-
fill gases with a wide range of compositions as typically found in the United States. The complete system
incorporates the landfill gas collection system, a fuel gas pretreatment system and a fuel cell energy conver-
sion system. In the fuel gas pretreatment section, the raw landfill gas is treated to remove contaminants to
a level suitable for the fuel cell energy conversion system. The fuel cell energy conversion system converts
the treated gas to electricity and useful heat.
Landfill gas collection systems are presently in use in over 100 MSW landfills in the United States. These
systems have been proven effective for the collection of landfill gas. Therefore these design and evaluation
studies were focused on the energy conversion concept.
3.2.1 Overall System Description
The commercial landfill gas to energy conversion system is illustrated in Figure 3-1. The fuel pretreatment
system has provisions for handling a wide range of gas contaminants. Multiple pretreatment modules can
be used to accommodate a wide range of landfill sizes. The wells and collection system collect the raw landfill
gas and deliver it at approximately ambient pressure to the gas pretreatment system. In the gas pretreatment
system the gas is treated to remove NMOCs including halogen and sulfur compounds. The pretreatment
system for the conceptual design is based upon a commercial system design operating at a landfill site in John-
ston, R.I. The system designed for this program has been modified to reflect the knowledge gained at that
site.
The commercial energy conversion system shown in Figure 3-1 consists of four fuel cell powerplants. These
power plants are designed to provide 200 kW output when operating on landfill gas with a heating value of
4.45 kcal per standard liter (500 Btu per standard cubic feet) and for accommodating higher contaminant con-
centrations. The output from the fuel cell is utility grade ac electric power. It can be transformed and put
inio the electric grid, used directly at nearby facilities, or used at the landfill itself. The powerplants are capa-
ble of recovering cogeneration heat for nearby use or rejecting it to air.
As configured in Figure 3-1, the commercial system can process approximately 504 standard cubic meters
per hour of landfill gas (mitigate 253 SCMD of methane) with minimum environmental impact in terms of
i iquids. solids, or air pollution. Details of the individual sub-elements in the energy conversion system follow
this discussion.
..^^^^^^^^^^^^^^^^^B
600 kW Fuel cell power plant
operating on landfill gas
Landfill site
office and
blower
system
Multiple
fuel ceil
power plants
FC30744
H941008
Figure J-7. Commercial Fuel Cell Landfili-Gas-to-Energy Conversion Concept
-12-
-------
International Fuel Cells
FCR-13524E
Fuel Pretreatment System
A block diagram of the landfill gas pretreatment system for the conceptual study is shown in Figure 3-2. The
fuel pretreatment system incorporates one non-regenerable step, plus two stages of refrigeration combined
with two regenerable adsorbent steps. The use of staged refrigeration provides tolerance to varying landfill
gas constituents. A non-regenerable carbon bed first removes hydrogen sulfide. The first stage condenser
removes the water content to a uniform dew point of approximately 1 ฐC, and removes some heavier hydrocar-
bons from the landfill gas. The first stage condenser provides flexibility to accommodate the varying landfill
characteristics by delivering a low dew point gas with a relatively narrow cut of hydrocarbons for the down-
stream beds in the pretreatment system. A regenerable dryer bed next reduces the dew point from 1 ฐC to less
than -45 ฐC, to prevent freezing in the second refrigeration step. The second refrigeration step enhances the
effectiveness of the activated carbon bed, which removes the remaining volatile organic compounds in the
landfill gas. This approach is more flexible than utilizing dry bed adsorbents alone and has built-in flexibility
for the wide range of contaminant concentrations which can exist from site to site and even within the single
site varying with time.
REGENERATION GAS
TO
FUEL
CELL
HEAT
RECOVERY
FOR
CONDENSATE
VAPORIZATION
AND
REGENERATION
GAS
HEATING
135224A
R981S04
Figure 3-2. Simplified Block Diagram of Commercial LFG Pretreaiment Sysrem
The two adsorbent beds are regenerated by using cleaned, heated gas from the process stream exit. Each ad-
sorbent step consists of two beds in parallel. In operation, one bed is adsorbing while the parallel bed is being
regenerated. The regeneration path and sequence are shown as dashed lines in Figure 3-2. A small portion
of the treated landfill gas (approximately 8%) is heated by regeneration with the incinerator gases and then
passed through the beds in the sequence shown. Figure 3-3 with its accompanying sample regeneration se-
quence shows the regeneration process in more detail. This system provides flexibility in the tailoring of the
regeneration of each bed. The exact sequencing, regeneration gas flow, and timing would be based on experi-
ence gained in the Phase II and III demonstrations and final design (bed sizing and material optimizations)
of the adsorbent beds for commercial applications. After exiting the final bed, the regeneration gas is fed
into the low NOX incinerator where it is combined with the vaporized condensates from the refrigeration pro-
cesses and the mixture is combusted to provide greater than 98 percent destruction of the NMOC's from the
raw landfill gas. The exhaust from the incinerator is essentially COi and water. The pretreatment system
design provides treated gas to the fuel power plant in an efficient, economic, and environmentally acceptable
manner.
Key features of the design and the related product benefits are summarized in Table 3-3. The system design
provides flexibility for operation on a wide range of landfill gas compositions, high thermal efficiency and
-13-
-------
International Fuel Cells FCR-13524E
low parasite power requirements. While the pretreatment system is based upon modification of an existing
system and it utilizes commercially available components, the process train and operating characteristics of
this design need to be validated by demonstration. Key demonstrations required include: the achievement
of the low total halide contaminant levels in the treated gas; effectiveness of the regeneration cycle as affected
by regeneration time and temperature; durability of the regenerable beds; and low environmental emissions.
The pretreatment system was analyzed to estimate the overall thermal efficiency, and internal electric power
requirements., and its maintenance characteristics. These characteristics are summarized in Table 3-4. The
estimated thermal efficiency is 92 percent with the balance of the thermal energy used for regeneration, vapor-
ization of condensates and incineration of the regeneration gases. Electric power is used for pumping the
gases and the refrigeration and is accounted for as a parasite power characteristic of the system. Maintenance
requirements consist of maintaining and adjusting controls and valves in the regeneration system replacement
of filter elements, replacement of fully regenerated spent bed materials on an annual basis and replacement
of the hydrogen sulfide removal bed on a periodic basis. The frequency is controlled by the hydrogen sulfide
content of the gas.
The environmental impact of the gas treatment system was evaluated. The impact of air emissions, liquids,
and solids disposal were considered. The incinerator is designed for greater than 98 percent destruction of
all NMOC's, and NOX emissions of less than 0.11 kg per 106 kcal of fuel consumed are expected. There is
no liquid effluent from the system since all condensates are vaporized and subsequently incinerated. Solid
disposal involves removing spent regenerable and non-regenerable bed materials at the factory and treatment
by an EPA approved processor for reclamation. The bed materials are routinely handled and processed by
qualified waste processors.
-14-
-------
International Fuel Cells
FCR-13524E
Regen
gas"j~
1
1
1
L
Regen
*. gas
heater
Bypass for
I
I
I
"1"*
1
Bypass
Dryer
bed
Bypass
r
1
_i
Activated
carbon
cooling
Staged Regeneration of Adsorbent Beds
8
Step
1
2
3
4
5
6
7
Duration
(hrs)
0.5
1.5
0.5
1.5
2.0
1.0
1.0
1
1
1
fc To
"" flare
FC32432
910409
Hour Sequencing
Mode
Heating
Heating
Heating
Heating
Cooling
Cooling
Cooling
Regen gas
heater
On
On
On
On
Bypass
Bypass
Bypass
Dryer
bed
Bypass
Regen |
Bypass
Bypass
Regen |
Bypass
Bypass
Activated
carbon bed
| Regen |
Regen
| Regen |
Bypass
| Regen |
Regen |
Bypass
Figure 3-3. Staged Regeneration of Adsorbent Beds and Sample Regeneration Sequence
Table 3-3. Key Features of Commercial Pretreatment System Conceptual Design
Design Feature
Product Benefit
4.8 x 104 Pa nominal operating pressure
Two refrigeration stages
Vaporization and incineration of liquid conden-
sates from refrigeration stages
Beds regenerated with heated clean fuel fol-
lowed by low NOX incineration
Recover heat from incineration for vaporization
of refrigeration condensate and heating clean
regeneration gas
Low pumping power
Handle wide range of landfill gases
Improve effectiveness of regenerable beds
No contaminated liquid effluents for disposal
No contaminated liquid effluent
High thermal efficiency
-15-
-------
International Fuel Cells FCR-13524E
Table 3-4. Gas Pretreatment System Projected Performance
Fuel pretreatment system efficiency (% of raw 92%
landfill gas delivered to fuel cell)
Parasite power requirement (% of fuel cell
electric power)
2%
Fuel Cell Power Plant
The commercial landfill gas energy conversion conceptual design incorporates four 200-kW fuel cell power
units. Since each of the four units in the concept is identical, this discussion will focus on the design issues
for a single 200-kW power unit.
A simplified functional schematic of the fuel cell power unit is shown in Figure 3-4. Major sections of the
system include the fuel processing system, fuel cell electrical conversion system and the thermal management
system. In the fuel processing section treated landfill gas is converted to hydrogen and COi for introduction
into the fuel cell stack. The fuel treatment process includes a low temperature fuel preprocessor to remove
the residual contaminants from the treated gas, a fuel reformer, and a low temperature shift converter where
the exhaust from the reformer is further processed to provide additional hydrogen and CC>2.
In the fuel cell stacks hydrogen from the process fuel stream is combined electrochemically with oxygen from
the air to produce dc electricity and byproduct water. The product water is recovered and used in the reformer.
The heat generated in the cell stack is removed to an external heat rejection system. This energy can be either
rejected to the ambient air or recovered for use by the customer. The dc power produced in a fuel cell stack
is converted to ac power in a power conditioning package not shown on the process schematic.
A preliminary design of a fuel cell power plant was established to identify the design requirements which
allow optimum operation on landfill gas. Three issues specific to landfill gas operation were identified which
reflect a departure from a design optimized for operation on natural gas. A primary issue is to protect the
fuel cell from sulfur and halide compounds not scrubbed from the gas in the fuel pretreatment system. An
absorbent bed was incorporated into the fuel cell fuel preprocessor design which contains both sulfur and
halide absorbent catalysts. A second issue is to provide mechanical components in the reactant gas supply
systems to accommodate the larger flow rates that result from use of dilute methane fuel. The third issue is
an increase in the heat rate of the power plant by approximately 10 percent above that anticipated from opera-
tion on natural gas. This is a result of the inefficiency of using the dilute methane fuel. The inefficiency
results in an increase in heat recoverable from the power plant. Because the effective fuel cost is relatively
low, this decrease in power plant efficiency will not have a significant impact on the overall power plant eco-
nomics.
The landfill gas power plant design provides a packaged, truck transportable, self-contained fuel cell power
plant with a continuous electrical rating of 200 kW. It is designed for automatic, unattended operation, and
can be remotely monitored. It can power electrical loads either in parallel with the utility grid or isolated from
the grid.
In summary, a landfill gas fueled power plant can be designed to provide 200 kW of electric output without
need for technology developments. The design would require selected components to increase reactant flow
rates with a minimum pressure drop. To implement the design would require non-recurring expenses for sys-
tem and component design, verification testing of the new components, and system testing to verify the power
plant performance and overall system integration. A thermodynamic analysis of the fuel cell power plant
optimized for operating on landfill gas was completed. The resulting performance of the landfill gas power
plant is compared to a power plant operating on natural gas in Table 3-5.
-16-
-------
International Fuel Cells
FCR-13524E
FUEL
PROCESSING
SYSTEM
LOS,TP?MP
PR|^
PROCESSOR
THERMAL-
MANAGEMENT
SYSTEM
Figure 3-4. Functional Schematic Fuel Cell Landfill Gas Power Unit
Table 3-5. Performance Comparison for Nominal 200 kW Output
Fuel
Electrical Efficiency (LHV) - %
Fuel Flow Rate (SUM)
Heat Rate (HHV) - kcal/kWhr
Available Heat - kcal/hr
Ambient Temperature for Fuel Water Recovery - ฐC
Startup Fuel
NATURAL GAS
POWER PLANT
Natural Gas
40.0
896
2,390
192,000
35
Natural Gas
LFG
POWER PLANT
Landfill Gas
36.4
1960
2,620
208,000
35
Landfill Gas
-17-
-------
International Fuel Cells
FCR-13524E
The estimated air emissions of the fuel cell power plant is provided in Table 3-6. The fuel cell air emissions
are low because gas contaminants which could become emissions are removed by the gas pretreater and fuel
preprocessor. The air emissions are significantly lower than other landfill gas conversion devices giving the
fuel cell power plant the potential for being the best available control technology for landfill gas methane
mitigation. Verification of these emission estimates will be a key element of the demonstration program.
Table 3-6. Estimated Fuel Cell Air Emissions
Emissions - kg/106 kcal
NOX
sox
Particulates
Smoke
CO
Total Hydrocarbons
LFG FUEL CELL
0.04 - 0.07
0.00005
0.000005
None
0.07-0.14
0.02 - 0.05
Overall System Performance
The commercial application of the concept to the market described previously was assessed. For the purpose
of the evaluation, a site capable of supporting four fuel cell power modules was selected. The site characteris-
tics assumed are the typical values discussed earlier. The site would produce approximately 12,200 standard
cubic meters of landfill gas per day. The gas contains approximately 50 percent methane by volume with
a heating value of 4.45 kcal/liter. The system is capable of supplying 784 kW of net electric power to the
grid and has an available thermal energy of 0.84 million kcal per hour. Overall system performance is outlined
in Figure 3-5.
-18-
-------
International Fuel Cells
FCR-13524E
FLARE GAS
EXHAUST
FUEL GAS
EXHAUST
LANDFILL
GAS
AIR
LFG
PRETREATMENT
SYSTEM
^ "^~-.
784 kWe
20.3 x106
KCAL/OAY
LANDFILL GAS IN
60BO SCMD CH4
6080 SCMD CO2
5.3 SCMD NMOC
WASTE FILTER
MEDIA
GAS PRETREATMENT
SYSTEM EFFLUENT
AIR EMISSIONS
15,100 SCMD N2, O2, H2OvCO2, Ar
0.11 SCMD NMOC
0.48 KG/DAY NOX
0.73 KG/DAY SO2
2.6 KG/DAY HCL
SOLID WASTE
25 KG/YR FILTER MEDIA
V
ADSORBENTS
FUEL CELL OUTPUT
784 kW AC ELECTRICITY
20.3 X 106 KCAL/DAY THERMAL ENERGY
AIR EMISSIONS
93,700 SCMD N2, H2Oyt CO2, O2, Ar
1.8 KG/DAY NO,
SOLID WASTE
68 KG/YR SULFUR AND HALIDE
ABSORBENTS (14 KG/YR SULFUR AND HALIDES)
13524-03
952104
Figure 3-5 Overall System Schematic and Performance Estimate for Fuel Cell LFG-to-Energy
Conversion System
Impact of Heating Value on System Performance
Heating value of the landfill gas can vary from site to site or at a given site with time The most significant
variation is a reduction in heating value from air intrusion into the landfill during energetic withdrawal and
collection of the gas. Although most of the oxygen in the air is consumed in the landfill, nitrogen content
of the gas increases thereby lowering the heating value of the gas.
Figure 3-6 shows the impact of changing the landfill gas heating value from 3.56 to 5.34 kcal/SL {kcal per
standard liter) on fuel cell power plant heat rate and power output. In general, the lower the heating value
of the gas the lower the power plant thermal efficiency will be (or otherwise stated, the higher the power plant
heat rate will be).
The power output of a fuel cell power plant optimized to operating on landfill gas with 50 percent methane
is shown in Figure 3-6. A reduction in methane content or heating value below 4.54 kcal/SL results in a loss
in energy input and power output. Above 4.54 kcal/SL the power plant automatic flow controls will self-ad-
just to maintain 200 kW output. Natural gas blending may be considered as a means to maintain the gas heat-
ing value above 4.54 kcal/SL at some sites.
-19-
-------
International Fuel Cells
FCR-13524E
200
WITH NATURAL GAS BLENDING
160
O
3 120
Q.
O
a. 80
oc
uu
O
Q.
40
2898
2772
i
I
2646
2520
Q.
DC
UJ
I
Q.
2394
3.56
4.54 5.34
LANDFILL GAS HIGHER HEATING VALUE - kcal/SL
13524-04
852104
Figure 3-6. Impact of Landfill Gas Heating Value on Power Plant Power Output and Heat Rate
3.2.2 Environmental and Economic Assessment of the Fuel Cell Energy Conversion System
The commercial application of the energy recovery concept to the market described previously was assessed
For the purpose of the evaluation, a site capable of supporting four fuel cell power plant units was selected
The site assumed characteristics, shown in Table 3-7, are the typical values discussed earlier. The site would
produce approximately 12,200 standard cubic meters of landfill gas per day. The gas contains approximately
50 percent methane and has a gas heating value of 4.54 kcal/SL.
At a minimum the method for mitigating the methane and NMOC from landfill gas is installation of a gas
collection system and flaring of the gas. The fuel cell energy conversion system provides the opportunity
for converting the methane in the landfill gas to useful energy. The baseline for the comparisons in this system
are the conventional option with flaring.
There are two significant differences between mitigation by flaring and mitigation with the fuel cell energy
conversion system. First, the fuel cell energy conversion system produces electric energy, thermal energy
and emissions offsets which can be used to generate revenues from the landfill gas mitigation system. Sec-
ondly, since the fuel cell converts methane to electricity more efficiently, it has lower emissions at the site
than competing options and provides significant emission offsets due to the reduction in emissions from the
electric utility which would otherwise be providing the energy. These differences are the basis for the asses-
sment of the energy conversion system discussed in this section. It should be noted that both the flare system
and the fuel cell system essentially eliminate all methane emissions from the landfill site and have a 98 per-
cent destruction of the non-methane organic compounds.
-20-
-------
International Fuel Cells
FCR-13524E
Table 3-7. Site Characteristics for Landfill Gas Assessment
Landfill Gas Generation Rate
12,200 SCMD
Bulk Constituents (vol %, dry)
Methane
CO2 and Other Inerts
50
50
Contaminants (PPMV)
Total Non-Methane Organic Hydrocarbons
Total Sulfur
Total Halides
Methyl Chloride
Vinyl Chloride
Gas Heating Value - kcal/SL
2700
21
132
14
7
4.54
Environmental Assessment
The analysis of the environmental impact shows that both the fuel cell and the flare system can be designed
to eliminate the methane and the non-organic methane compounds from the landfill gas system. For the ex-
ample site considered, the methane elimination is essentially complete for both systems and 98 percent of
the NMOC are destroyed. Trace amounts of SOX and NOX will be emitted in each case. With the fuel cell
system, however, significant reductions of NOX and SOX will be achieved due to the fuel cell energy genera-
tion. This analysis assumes an 80 percent capacity factor for the fuel cell and offsetting emissions from elec-
tric utility power generation using a coal-fired plant meeting New Source Performance Standards. For the
example site, the fuel cell energy conversion system provides 5.6 million kWhr of electricity per year, with
a net reduction of 32.0 Mg per year of NOX and 15.2 Mg per year of SOX from reduced coal use. These reduc-
tions can be used as environmental offsets, particularly in critical areas such as California or other locations
with stringent environmental requirements.
The environmental impact of application of the fuel cell concept to the potential market is shown in Table
3-8. The data show that both the flare and the fuel cell mitigate methane and NMOC, under the proposed
standards and guidelines1. However, the flare merely converts these emissions to CO2, acid rain, and other
unhealthy pollutants. The fuel cell can provide a net reduction in global pollution by offsetting energy pro-
duction from coal.
Table 3-8. Emissions Impact of Fuel Cell Energy Recovery from Landfill Gas
Abatement
Technology
Venting
Only
Flare
Fuel Cell
Global Warming
Methane
(Mg/Yr)
1.8 x 107
0
0
NMOC
(Mg/Yr)
510,000
10,200
10,200
CO2
(Mg/Yr)
4.94 x 107
-6.45 x 107
Acid Rain and Health
SO2
(Mg/Yr)
2,972
-535,000
NOX
(Mg/Yr)
29,720
-259,000
CO
(Mg/Yr)
14,860
-8,620
-21-
-------
International Fuel Cells FCR-13524E
Economically the fuel cell energy system has the potential for deriving revenues from electric sales, thermal
sales, and emission offsets credits. These revenues can be used to offset the investment cost associated with
gas collection, gas pretreatment, and fuel cell power units. The level of these revenues depends upon the value
of the electricity, the amount and value of the heat used, and the value of the emissions offsets.
Economic Assessment Results
The fuel cell energy system has the potential for deriving revenues from electric sales, thermal sales and emis-
sion offsets credits. These revenues can be used to offset the investment cost associated with the gas collec-
tion, gas pretreatment and the fuel cell power units. The level of these revenues depends upon the value of
the electricity, the amount and value of the heat used and the value of the emissions offsets.
Electric rates vary considerably with geographic location and the purchaser of the electric energy. Commer-
cial rates are applicable where the electricity can be used at the landfill or in nearby commercial facilities.
Commercial rates vary from a high of 13.68 cents per kWhr to a low 2.71 cents per kWhr. The median rate
in the United States is approximately 7 cents per kWhr. The rates charged to industry are generally lower
and are closer to the fully burdened avoided cost for the utility. These rates range from 10.0 cents per kWhr
to a low of 1.64 cents per kWhr with the mean value of approximately 5 cents per kWhr. In general, both
the commercial and industrial rates are higher in locations with high population density and/or with air emis-
sions problems. These locations are idea! for the use of the fuel cell energy conversion system with its favor-
able environmental impact. Since the rates vary considerably, the analysis in this section is done on a
parametric basis for a wide range of electric rates.
The fuel cell energy conversion system was evaluated to establish the net revenues or costs for processing
landfill gas to mitigate methane emissions. For the purposes of the analysis it was assumed that the fuel cell
energy conversion system and the flare system would have an overall annual capacity factor of 80 percent.
For this analysis, two levels of fuel cell installed cost were considered. The lower level, $ 1500/kW represents
a fully mature cost when the power plant has been accepted into the marketplace and is routinely produced
in large quantities. The upper level, $3000/kW installed, represents a price level when the power plant is
being introduced into the marketplace, and is produced on a moderate and continuous basis. In addition to
the fuel cell costs, a GPU installed cost of $190/kW and gas collection system cost of $310/kW are included
in the overall system cost. Operating and maintenance costs are 0.4tf/kWHr for the GPU and 1.5
-------
International Fuel Cells
FCR-13524E
Figure 3-8 shows the fuel cell revenues for situations without heat recovery. Although the net revenues are
somewhat decreased, the results and areas of competitiveness are similar to those noted for Figure 3-7.
FUEL CELL INSTALLED COST
3000
U to
C: UJ
~ X 2000
81
DC -J
U. u.
) O
li
UJ O
DC oo"
u ft
z
1000
-1000
-2000
MATURE
PRODUCT/
COST
ENTRY LEVEL
COST
FUEL CELL REVENUES FROM
ELECTRICITY
EMISSION CREDITS
THERMAL RECOVERY
\
I
I
FLAR*fฃฐNOMIC
OPTION
I
\
COLLECTION
AND FLARE
I
2.0 4.0 6.0 8.0 10.0 12.0 14.0
VALUE RECEIVED FOR FUEL CELL ELECTRICITY ~ c/kWr
13524-05
R070212
Figure 3-7. Comparison of Fuel Cell to Flare for Methane Mitigation Assuming Electric Revenues.
Emission Credits and Thermal Recovery
Z
o Q
F UJ
< CO
CD (O
F S
ป8
UJ O.
Z en
^l ^L
X O
tJJ
o <
cc -J
LL LL
to ฐ
UJ S
i^
>i
UJ O
CC to
UJ *ป
z
3000
2000
1000
-1000
-2000
FUEL CELL INSTALLED COST
MATURE
PRODUCT
COST
FUEL CELL
ECONOMIC OPTION
ENTRY LEVEL
COST
FUEL CELL REVENUES FROM
ELECTRICITY
EMISSION CREDITS
\
FLARE ECONOMIC
OPTION
GAS COLLECTION
AND FLARE
2.0 4.0 6.0 8.0 10.0 12.0 14.0
VALUE RECEIVED FOR FUEL CELL ELECTRICITY ~e/kWhr
13S24-08
952104
Figure 3-8. Comparison of Fuel Cell to Flare for Methane Mitigation Assuming Electric Revenues and
Emission Credits
-23-
-------
International Fuel Cells
FCR-13524E
Figure 3-9 shows the fuel cell revenues for the most stringent application situation. In this case, the fuel cell
receives revenues only from the sale of electricity. Although the emissions are lower from the fuel cell, no
specific credit or value is attached to them for this example. Under these conditions the fuel cell is still the
economic choice for most locations at $1500/kW. At $3000/kW it is still economical in those areas where
the value of electricity is nine cents per kWhr or higher. This would primarily be areas such as California,
New York, and parts of New England.
This analysis indicates that there is substantial market opportunity for the fuel cell energy conversion system.
At market introduction prices, the power plant would be applicable in locations with high electric rates and
situations where ah- emissions are quite severe. There are many areas of the country which have these charac-
teristics and they are increasing with time, thus indicating an ever increasing market opportunity for the fuel
cell power plants. These options were evaluated against the flare option which does not produce any useful
energy. There are other energy conversion systems which could produce electric and/or thermal energies.
Comparisons to these options are discussed in the following section.
a
H IU
i 3
_ o
I ง
ec -J
LL u.
t/J O
UJ J?
=> o
z to
UJ p
> o
HI O.
ff* CO
3000
2000
1000
-1000
-2000
FUEL CELL INSTALLED COST
MATURE
PRODUCT,
COST
ENTRY LEVEL
COST
FUEL CELL REVENUES FROM
ELECTRICITY
FUEL CELL
ECONOMIC OPTION
\
FLARE ECONOMIC
OPTION
I
I
GAS COLLECTION
AND FLARE
I
2.0 4.0 6.0 8.0 10.0 12.0 14.0
VALUE RECEIVED FOR FUEL CELL ELECTRICITY ~ C/kWhr
13524-07
952104
Figure 3-9. Comparison of Fuel Cell to Flare for Methane Mitigation Assuming Electric Revenues Only
Comparison With Other Energy Conversion Options
The internal combustion engine and the gas turbine engine have been suggested as competing options for
methane mitigation at landfill sites. For the landfill size selected for this analysis, the internal combustion
engine is more effective than the gas turbine option for clean-up. This is used as the basis for the comparisons
in this section. The internal combustion engine can provide both heat and electric energy while consuming
the methane at the landfill gas site. With the present state-of-the-art technology, however a lean-bum internal
combustion engine has higher levels of NOx unless special precautions are taken to clean-up the exhaust.
For this analysis, two cases are considered. The first case assumes there is no clean-up of the exhaust from
the lean-bum internal combustion engine and the second assumes that the exhaust is cleaned with selective
catalytic reduction (SCR). Since the SCR employs catalyst in the clean-up system, the landfill gas will have
to be pretreated in a manner similar to the fuel cell system. For those cases with the SCR clean-up system,
a pretreatment system has also been included as part of the total system cost.
-24-
-------
International Fuel Cells
FCR-13524E
Figure 3-10 shows the results of the economic analysis for the fuel cell system and the internal combustion
engine system. Since both can provide electricity, me comparison between the systems are based on the cost
of electricity generated from the energy conversion system with appropriate credit for thermal sales and/or
emission offsets. For the case where the SCR is employed to clean-up the engine exhaust, the fuel cell power
plant is competitive with installed prices on the order of $3000/kW. If no exhaust clean-up is required for
the internal combustion engines, then the fuel cell is competitive at the fully mature price of $1500/kW. In
this latter case, however, the operation of the internal combustion engine at the landfill site would be quite
dirty and significant amounts of NOx would be added to the ambient air. For many locations where the fuel
cell would be considered, such as California or other high emissions areas, the no exhaust clean-up option
may not be available. Consequently, the fuel cell option would be fully competitive with the internal combus-
tion engine option for most cases where on-site clean-up of the internal combustion engine is required.
10.0
C- B.O
E 6.0
O
I
LJ
U.
O
to
O
4.0
ELECTRICITY SALES
THERMAL RECOVERY
EMISSIONS OFFSETS
1500S/KW
WITH
SCR
EXHAUST
CLEAN-UP
NO
EXHAUST
CLEAN-UP
FUEL CELL
ENERGY CONV
SYSTEM
LEAN-BURN INTERNAL
COMBUSTION ENGINE
ENERGYCONV.
SYSTEM
13524-03
9S2104
Figure 3-10. Comparison of Fuel Cell to Internal Combustion Engine Energy Conversion System
Based on the analysis of the flare and other energy conversion options, the fuel cell power plant is fully com-
petitive in all situations in the mature production situation. For initial power plant applications with limited
lot production, the fuel cell powerplant is competitive in areas with high electric rates and/or severe emissions
restrictions at the local landfill site.
Conclusions
Based on the environmental and economic evaluation of the commercial fuel cell energy system, the follow-
ing conclusions can be made:
The fuel cell landfill gas to energy conversion system provides net reduction in total emissions
while simultaneously mitigating the methane from the landfill gas.
With the initial product prices, fuel cells will be competitive in landfill sites located in high elec-
tric cost areas in sites with average commercial rates; where heat can be utilized or where there
is a credit for the environmental reductions from the fuel cell energy conversion system.
When the projected mature product price is achieved, fuel cells will be competitive for most
application scenarios. In many situations, fuel cells will provide net revenues to the owners of
the operating landfills. This could, in the long term, result in methane mitigation without addi-
tional cost of any son to the ultimate consumer.
-25-
-------
International Fuel Cells FCR-13524E
3.2.3 Critical Issues
This section summarizes the key marketing and technical issues that must be resolved to verify the commer-
cial feasibility of the fuel cell landfill gas to energy conversion concept. Resolution of some of these issues
will come with the recognition of the long term economic value of the fuel cell in mitigating landfill methane
and NMOC emissions while significantly lowering secondary emissions and offsetting the air emissions
from electric utility generators. Resolution of other issues will be achieved with the design and successful
demonstration of the pretreatment system and fuel cell on landfill gas. The following marketing and technical
issues need to be resolved:
Marketing Issues
Market Entry at Initial Product Capital Cost - Market acceptance of the fuel cell energy recovery
concept must be achieved by entry into markets with the highest electric rates or strictest emis-
sion controls. Federal incentives such as; low cost financing, emission credits, etc. can hasten
acceptance of the concept.
Limited Electric Revenues - Electric utility avoided cost rates are impeding energy recovery
from sources such as landfill gas. Allowing revenues based upon the local commercial or indus-
trial rates, or fully burdened avoided costs would encourage energy recovery and thus achieve
the desired environmental impact.
Available Uggsgf TherroalEnejgy - Fuel cell revenues increase with the sale of thermal energy.
Identification of thermal loads near landfills or arrangements to locate the fuel cell at cogenera-
tion sites near landfills (with gas pretreatment located at landfill) would improve fuel cell market
competitiveness.
Technical Issues
Verification of an Effective Pretreatment System - The prelreatment system design must eco-
nomically treat landfill gas to meet the long term sulfur and halide limits required by the fuel
cell. This demonstration program will verify the key process elements and steps of the commer-
cial system. The system and its elements must be optimized to produce a cost effective commer-
cial pretreatment system.
Demonstration of Low Emissions - The demonstrator design and actual demonstration will
verify the low emission capability of the commercial fuel cell landfill-gas-to-energy concept.
This includes air emissions from both the fuel cell and pretreatment system as well as solid and
liquid effluents projected for the commercial system.
Demonstrate Overall System Operability. Durability and Reliability - A successful demonstra-
tion will allow projection of a low operating cost component of the methane mitigation life cycle
cost for the commercial system. This includes trouble-free unattended operation and minimal
degradation (durability) of the regenerable beds.
-26-
-------
International Fuel Cells FCR-
4.0 DEMONSTRATION TEST DESIGN
This section describes the site specific process and engineering design of the gas pretreatment unit designed
for the landfill gas application, plus the PC25 A, 200-kW phosphoric acid fuel cell power plant, and the land-
fill-gas-to-energy site which was selected in Phase I. The PC25 A fuel cell is uniquely suited to this applica-
tion due to its low secondary emissions: typically 0.5 ppmV NOX, 1.1 ppmV carbon monoxide; and 0.03
ppmV non-methane hydrocarbons (all measured at 15 percent oxygen on a dry gas basis using natural gas
fuels). Section 4.1 describes the development of the gas pretreatment process from a conceptual design in
Phase I, to a complete, detailed mechanical and process design including field modifications in Phase II. In
Section 4.2 the PC25 A power plant design modifications to permit operation on landfill gas up to a nominal
140 kW rating are described. Section 4.3 describes the overall site specific process design for the demonstra-
tion including the gas pretreatment system and its integration to the fuel cell power plant system. Section
4.4, Site Specific Engineering Design, summarizes the details of the site location and construction for the
demonstration equipment.
4.1 Select Landfill Site
The objective of the site selection effort was to select the best available site for demonstrating the re-
covery of energy from landfills using fuel cells. The approach was to establish site selection criteria
from the conceptual design of the commercial product; apply these criteria to potential landfill sites
identified by Pacific Energy, and then downselect and rank these sites according to the criteria. Based
on this evaluation, Pacific Energy's Penrose Power Station in Sun Valley, California, was selected as
the site for the demonstration.
4.1.1 Site Selection Criteria
Two major site selection criteria were established for selecting the demonstration site. These criteria
are:
(1) That the site be representative of U.S. landfills, so that the demonstration will be relevant to a large
portion of the U.S. market; and
(2) That the site be suitable, with existing and reliable gas supplies and facilities available.
The first selection criterion, that the site be representative, requires that the landfill gas available at
this site be typical of the majority of sites across the United States in major gas composition, heating
value and contaminants. Equally important, is that the local codes and regulations be sufficiently de-
manding that a successful siting and demonstration at the selected site would be readily accepted at
most if at not all potential sites across the United States.
The suitability criterion relates to. the practicality and expense of conducting the demonstration at a
candidate site. Specifically, a suitable site should have all required facilities in place including a prov-
en landfill gas collection system, natural gas supply, permits, contracts for equipment and sale of elec-
tricity, and plenty of space available for the demonstration. Most importantly, the selected site should
have an excess supply of landfill gas available for the demonstration during the proposed period of the
demonstration gas pretreatment test in 1993 and the fuel cell demonstration in 1994 and 1995.
4.1.2 Characteristics of Candidate Sites and Selection
The candidate site selection was based upon the twelve active projects which Pacific Energy currently
manages, as shown in Table 4-1. A preliminary assessment conducted by Pacific Energy identified four
of these sites as potential candidates for the EPA demonstration. These sites are: the Oxnard Station
in Oxnard, California; the Penrose Station in Sun Valley, California; the Toyon Station in Los Angeles,
California; and the Otay Station in San Diego, California. Tables 4-2 and 4-3 show how these four
candidate sites were assessed against the detailed sub-criteria which were developed in the program.
The two leading candidate sites that emerged were Penrose and the Toyon Canyon site because of their
-27-
-------
International Fuel Cells
FCR-13524E
close proximity to the Pacific Energy and Southern California Gas facilities, and their location within
the South Coast Air Quality Management District. Oxnard and Otay are less attractive due to their
greater distance from Los Angeles, and lack of natural gas service.
The Toyon site was eliminated when it was determined that there was insufficient excess landfill gas
available year-round for the demonstration, particularly during the summer months. This left the Pen-
rose site as the best site for the demonstration.
The Penrose site is representative of U.S. landfills. Landfill gas is provided from four separate land-
fills with typical levels of contaminants and gas heating value. This site is also regulated by the South
Coast Air Quality Management District which is nationally recognized for strict environmental regu-
lations, so that a successful demonstration in this area is likely to be accepted by other localities within
the United States.
The Penrose site is entirely suitable for the demonstration. All required facilities are already in place
including: a proven landfill gas collection system, natural gas supply permits, contracts for equipment
and saleable electricity, and more than adequate space available for the demonstration. With the re-
cent tie-in to a fourth landfill, there is an excess of landfill gas available to provide the 2350 SL/M re-
quired for the demonstration. The site is readily accessible within a half-hour of the main facilities of
the Pacific Energy and Southern California Gas. Support facilities and personnel are already in place.
Table 4-1. Pacific Energy Landfill Gas Sites
Landfill Gas Projects
Upland Pwr Sta.
Oxnard Pwr. Sta.
Penrose Pwr. Sta.
Toyon Pwr. Sta.
Gude Pwr. Sta.
Bakersfield Pwr. Sta.
Stockton Pwr. Sta.
Lompoc Pwr. Sta.
Crazy Horse Pwr. Sta.
Santa Clara Pwr. Sta.
Oiay Pwr. Sta.
Bonsai 1 Pwr. Sta.
Location
Upland CA
Oxnard CA
Sun Valley CA
Los Angeles CA
Rockville MD
Bakersfield CA
Stockton CA
Lompoc CA
Salinas CA
Santa Clara CA
San Diego CA
San Diego CA
type
Elec. Pwr.
Elec. Pwr.
Elec. Pwr.
Elec. Pwr.
Elec. Pwr.
Elec. Pwr.
Elec. Pwr.
Elec. Pwr.
Elec. Pwr.
Elec. Pwr.
Elec. Pwr.
Elec. Pwr.
MWe
0.6
3.7
9.3
9.3
3.0
1.8
0.8
0.6
1.4
1.5
1.9
L5
Power
Purchaser
SCE
SCE
SCE
SCE
PEPCO
PG&E
PG&E
PG&E
PG&E
PG&E
SDG&E
SDG&E
35.4
-28-
-------
International Fuel Cells
FCR-13524E
Table 4-2. Assessment of Candidate Sites vs. Evaluation Criteria
1. Select site representative of U. S.
Landfills so demonstration will be
relevant to a large portion of the U. S.
Market (compare landfill to range for
U.S. Landfills.
Gas Heat Content (Btu/Ft3 HHV)
CH4
Gas Diluents
%02
%N2
%C02
Gas Contaminants
Toial NMOC (ppm)
Total Sulfur (ppm)
Total Halides (ppm)
Local codes and regulators - to dem-
onstration at this site likely to be ac-
cepted at other localities (yes/no)
Name of local regulatory agency
Geographic/market potential - is
demonstrated located in area likely
IO support commercial fuel cell
LFG-to-energy (yes/no)
Size of existing Landfill Power Plant
2. Suitability of site for demonstration
test
Multiple landfill gas sources
Site, setup for LFG-to-energy
Collection system-in-place (yes/no)
Gas contracted (yes/no)
Permit in place (yes/no)
Power contracted in place (yes/no)
Space available for demonstration
(yes/no)
Excess gas availability for demon-
stration (at least 84 SCFM)
Natural gas service available
Opportunity to demonstrate thermal
utilization (Note: Little potential use
at any site)
Accessibility for workers and visi-
tors (Ex/Good'Poor)
Support facilities
Maintenance
Phone/Office
Sanitary
Support personnel available (num-
ber) at site
Site Ownership
Site Aesthetics
Landfill Status
Typical
USA
500 Btu/Ft-1
50%
< 1%
<2%
50%
274
21
132
Penrose
440
44
<)%
4%
54%
130-474
150
78-95
Yes
SCAQMD
Yes
8.9 MW
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
1. Plant Office
Excellent
Yes
Yes
Yes
5
Private Industrial
Area
Fair
3 Closed
1 Open
Toyon
470 - 550
47-54
< 1%
7%
39%
36-130
14-24
7-58
Yes
SCAQMD
Yes
8.9 MW
No
Yes
Yes
Yes
Yes
Yes
No
No
Yes
1. Plant Office
Excellent
Yes
Yes
Yes
S
CityofLA.
Dept. of Parks &
Rec.
(Griffith Park)
Excellent
1 Closed
Oxnard
550 - 570
54
<1%
6%
39%
36-50
Not available
Not available
Yes
Ventura APCD
Yes
3.7 MW
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
1. Plant Office
2. Hotel
3. Club House
Excellent
Yes
Yes
Yes
2
City of Oakland
Excellent
2 Closed
1 Open
Otay
550
54
0.2%
3%
43%
Not available
27-63
Not available
Yes
San Diego APCD
Yes
1.9MW
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
1 . Plant Office
Poor
Yes
Yes
(Portable Only)
1
County of San
Diego
Good
1 Open
-29-
-------
Internationa] Fuel Cells
FCR-13524E
Table 4-3. Supplemental Landfill Data for Candidate Sites
Number of landfills
serving surnames
Ownership of Land-
fills
Distance from Down-
town Los Angeles
Siie Description
Ownership of Pro-
posed Fuel Cell Site
Status of Landfills -
Open (Ol/Closed (C)
Penrose
l.Penrose
2.Shelton Arleta
3. Bradley
4. Tuxford
1 . Private
2. City of L. A.
3. Private
4. Private
IS miles (northwest) Sun
Valley/Burbank
Old Industrial
(.Private Industrial
2. Pacific Energy
l.Penrose (C)
2. Shelton-Arleta (C)
3. Bradley (C)
4. Tuxford (C)
Toyon
l.Toyon
1. City of L. A.
7 miles (northwest)
Griffith Park
Woods - View of Moun-
tains
1 . City of L.A. Dept. of
Parks (Griffith)
l.Toyon (C)
Oxnard
I.Santa Clara
2. Ventura Coastal
3.Ballard
1. City of Oxnard
2. Sanitation Dist.
3. Private
70 miles (west) City of
Oxnard
River, New Hotel, Golf
Clubhouse. Homes
1. City of Oxnard
2. Ventura County
I.Santa Clara (C)
2. Ventura (C)
3.Bailard (O)
Otay
l.Otay
1. County of San Diego
100 miles (south) City of
Chula Vista
Hills, Light Industrial
1. City of San Diego
l.Otay (O)
4.1.3 Description of Selected Site
A detailed description of the selected site is given in Figures 4-1 and 4-2. These figures, provided by
Pacific Energy, describe the location and description of the existing landfill gas to energy conversion
equipment which consists of five 9.375 megawatt internal combustion engine generator sets. This site
presently produces 8.9 megawatts of net power to the electrical grid.
Figure 4-3 shows an aerial photograph of the Penrose Station which is located off Tujunga Avenue
along the edge of the Penrose landfill. Site potential locations for the demonstration are identified on
the aerial photograph. Site location number two was selected based upon its close proximity to the
existing power plant station, ease of access to Tujunga Avenue, plus availability of virgin soil for laying
foundations for the gas pretreatment and fuel cell equipment.
-30-
-------
International Fuel Cells
FCR-13524E
PACIFIC LIGHTING ENERGY SYSTEMS
LANDFILL GAS
TO ELECTRIC
POWER
Situated m Sun Valley, an indus-
trial residential area in Los
Angeles San Fernando Valley the
Penrose power station like its twin
at Toyon Canyon began opera-
tion in 1985 as one of the world s
largest landfill gas-to-electric
power plants The gas from this
relatively deep landfill fuels five
Clean Burn internal combustion
engines to produce a maximum
8 9 megawatts (MW) of electrical
power which is sold to Southern
California Edison This alternative
energy resource can serve the
electrical needs of an estimated
8 900 homes ana help conserve
Our natural energy resources by
saving the equivalent of up to
115 000 barrels of Oil per year In
addition the proiect produces
risk-free revenue for the city pays
local state ana federal taxes and
provides environmental and
safety benefits by reducing gas
migration and surface gas
emissions
Landfill Description
The Penrose sanitary landfill
owned by Los Angeles By-ProOucts
inc contains an estimated 9 mil-
lion tons of non-hazardous munic-
ipal waste The landfill opened in
Penrose Power Station
Penrose Plant Supplies Alternative Energy
to Southern California Power Grid
Figure 4-1. Penrose Plant Supplies Alternative Energy to Southern California Power Grid
(Courtesy of Pacific Energy)
-31-
-------
International Fuel Cells
FCR-13524E
Landfill Gas to Electric Power
Penrose Power Station
IGasWelKiBS)
I and Collection Syste
Eleclncal Transformer
(1)
Electrical Oermialors
(5)!
Gas Compiessors I ! Engines
I6J j(b)
Gas Proliltcr I o,. -., ,-.. ,-\ _ .-. ,.-
m\ 1,1 111.! :!- ,!_-
'960 ana closed m 1983 It covers 72 acres
ana has an average depth of 200 feel.
maKirig it one of the Oeeoest of Pacific
Lighting Energy Systems (PLES) landfill
proiects The landfill is located about fifteen
mnes northwest of aowntown Los Angeles
m Sun valley ana one mile west of the
Goiaen Slate freeway (Route 5) at Penrose
Avenue
Power Station Description
The cower station and gas collection sys-
tems were designed ana developed
svr-a caiea and are operated by PLES
Construction work was completed in 10
months ana start up began in December
985
The gas collection system consists of 85
wens of which 55 are single pipe wells and
30 are aupiex wells containing two pipes .
The aupiex wells recover gas from the
miooomt ana bottom of the landfill Each
wen p.pe M" or 6"pvc) is slotted on the bot-
tom inira to recover the gas produced and
nas a butterfly valve installed on top to con-
t'oi aas How The wens are interconnected
PV a surtace ana subsurface pipeline sys-
tei~ vvHicn tueis the power station locateO in
:ne nor:neast corner of the site
'.V.thin me station six 150 horsepower
^oiO'-anven reciprocating compressors
cra.\ Tie aas from me landfill and through a
v.o-stage oil oath prelilter at 20" to40"vac-
H^T- The gas trom me compressors at
90 - '00 usig passes through a pulsation
r ana two coalescing (liters and
-------
International Fuel Cells
FCR-13524E
Figure 4-3. Fuel Cell Site Optit
(Courtesy of Pacific Energy)
>tions
-33-
-------
International Fuel Cells
FCR-13524E
4.2 Landfill Gas Pretreatment Unit Process Design and Description
The process design of the landfill gas pretreatment unit cleaning process is dictated by the final gas purity
requirements for the fuel cell, the composition of the incoming landfill gas and its complex mixture of trace
contaminants, and the requirement that the gas cleanup process be capable of handling variations in inlet gas
composition. The fuel cell gas quality must be essentially free from all sulfur and halogen contaminants so
as to consist primarily of a mixture of methane, nitrogen, oxygen, and carbon dioxide. The landfill gas pre-
treatment unit design specification (See Appendix B, Attachment A, pB-19 to B-29) allows a maximum exit
sulfur level of 3 ppmV and maximum halogen level of 3 ppmV. This level of sulfur can be successfully re-
moved by the fuel cell power plant internal fuel cleanup subsystem while the halogen can be removed with
the addition of an optional halide guard to the existing system. The EPA field test results described in Section
5.3 show landfill gas pretreatment unit contaminant cleanup levels far better than the specification require-
ments. Raw landfill gas trace contaminants and their concentration levels used as the original basis for the
landfill gas pretreatment unit process design are shown in Table 4-4. The hydrocarbon and contaminant spe-
cies in raw landfill gas consists of a mixture of saturated hydrocarbons, aromatics, halogenated hydrocarbons,
hydrogen sulfide and organic sulfide gases.
Table 4-4. Raw Landfill Gas Contaminants and Concentrations at Penrose Test Site
(Original Pre-EPA Program Data)
Landfill Gas Trace Contaminants
Hydrocarbons
Isobutane
Isopentane
n-Pentane
Hexane
Octane
Aromatjcs
Benzene
Ethylbenzene
Chlorobenzene
Toluene
Xylenes
Styrene
Halogenated Hydrocarbons
Dichloroethene
Dichloroethane
Methylene Chloride
Cis-1, 2-Dichloroethene
Trichlorofiuoroethane
Trichloroethylene
Tetrachlorethylene
Vinyl Chloride
Sulfides
hydrogen Sulfide
Methyl Mercaptan
ithyl Mercaptan
Dimethyl Sulfide
Dimethyl Disulfide
Raw Gas Concentration Level
(ppm - by volume)
95
963
198
297
81
2
13
1
35
22
0.5
3
3
12
5
0.6
70
6
1.4
103
5
5
8
0.02
-34-
-------
International Fuel Cells
FCR-13524E
4.2.1 Process Operation
A simplified schematic of the landfill gas pretreatment system is shown in Figure 4-4. The process consists
of ambient temperature H2S removal followed by cooling, condensation, drying, further cooling, hydrocar-
bon removal and final filtration. The process is designed to remove the hydrogen sulfide contamination and
water vapor in the early process steps, so that final polishing can be accomplished over an activated carbon
bed which is maintained at a constant low temperature, to insure consistent high trace contamination removal.
This design makes the process relatively insensitive to changes in the gas inlet gas concentration with time,
and thus makes it an excellent candidate for landfills. The landfill gas pretreatment system is comprised of
the following 3 major subsystems:
Clean Gas Production Process
Regeneration Process
Refrigeration Process
i^>
H2S
REMOVAL
=>
COOLER
CONDENSER
3>
DRYER
BED
=>
LOW
TEMPERATURE
COOLER
T ADSORPTION AC
CONDENSATION OF WATER. HYDRO
nc WATFD ORGANIC SULFUR SULFU
HYDROCARBONS AND HALOGEN
HYDROCARBONS COMPOUNDS
260*C
REGENERATION
ADSORPTION OF REMAINING
HYDROCARBONS INCLUDING ORGANIC
SULFUR AND HALOGEN COMPOUNDS
2WC
REGENERATION
TO -ซ
FLARE
WATER
DESORPTION
FLARE
H/C
DESORPTION
T
TOO si ru
REGENERATION
GAS
Figure 4-4. Landfill Gas Pretreatment Unit System
4.2.1.1 Clean Gas Production Process - The purification process is represented in a block flow diagram in
Figure 4-5 . The process operates on raw landfill gas which is regulated down to 1.52x1 05 Pa from the Pen-
rose plant compressor. This process incorporates HaS removal, refrigerated cooling, and condensation to
remove water, adsorption drying, cooling, and hydrocarbon adsorption process units to remove contaminants
from the landfill gas.
The H2S removal bed reacts H2$ with C>2 found in the landfill gas to produce elemental sulfur. This bed con-
tains 1 19 liters (43 cm diameter x 81 cm deep) of activated carbon impregnated with potassium hydroxide,
from Westates Carbon. This bed is non-regenerable and is replaced periodically. The first stage cooler con-
denser operates at approximately +2ฐC and the second stage cooler operates at -28ฐC.
-35-
-------
International Fuel Cells
FCR-13524E
i
RAW LANDFILL GAS AT1.52x105 Pa
NON-REGENERABLE H2S REMOVAL BED
32" C
FIRST STAGE COOLER CONDENSER
2'C
FIRST STAGE
LIQUID COALESCING SEPARATOR
REGENEBABLE DRYER BEDS (2)
TO10ฐC
SECOND STAGE REFRIGERATION COOLER
28ฐC
SECOND STAGE
LIQUID COALESCING SEPARATOR (NOT REQUIRED)
-28ฐC
REGENERABLE HYDROCARBON ADSORPTION BEDS (2)
I
-26ปC
PARTICULATE FILTER
I
26ฐC
AMBIENT AIR FINNED
TUBE HEAT EXCHANGER
REGENERATION
PROCESS
1.24 x 10s Pa
TO FUEL CELL SUPPLY REGULATOR
EXISTING
CONDENSATE
SYSTEM
13574-13
981701
Figure 4-5. Gas Purification Process
The first stage cooler removes water, some heavy hydrocarbons, and sulfides which are discharged as conden-
sate to the Penrose plant's existing gas condensate pretreatment system. Since the demonstration landfill gas
pretreatment unit (GPU) operates on a small slip stream from the Penrose site compressor and gas cooler,
-36-
-------
International Fuel Cells FCR-13524E
some of the water and heavy hydrocarbons species are removed prior to the GPU. Most of the contaminant
halogen and sulfur species are lighter and remain in the landfill gas to be treated in the pretreatment unit.
All remaining water in the landfill gas, as well as some sulfur and halogen compounds, are removed in a re-
generable dryer bed which has a high capacity for adsorbing the remaining water vapor in the landfill gas.
The bed is 119 liters total volume (43 cm diameter by 81 cm deep) filled with 71 liters of Alcoa F200 aluminia,
followed by 48 liters of Davidson 3 A mole sieve. There are two dryer beds so that one is always operational
while the other is being regenerated. The dry landfill gas is then fed to the second stage cooler. This cooler
can be operated as low as -32ฐC and potentially can condense out heavy hydrocarbons if present at high
enough concentrations. In addition the second stage cooler reduces the temperature of the carbon bed there-
fore enhancing its adsorption performance7. The downstream hydrocarbon adsorption unit whose tempera-
ture is controlled by the second stage cooler is conservatively sized to remove all heavy hydrocarbon, sulfur
and halogen contaminant species in the landfill gas. This unit consists of two beds, each containing 119 liters
of activated carbon (Bameby and Sutcliffe, type 209C) so that one is always operational while the other is
being regenerated. Both the regenerable dryer and hydrocarbon removal beds operate on a nominal 16 hour
cycle with each set of beds operating in the adsorption mode for eight hours and regeneration mode for eight
hours. The gas then passes through a paniculate filter and is warmed indirectly by an ambient air finned tube
heat exchanger to insure a fuel inlet above 0ฐC before being fed to die fuel cell unit. The GPU process operat-
ing pressure is nominally 1.38 x 105 Pa with minimal pressure loss across the equipment. A final regulator
reduces the landfill gas pressure to the fuel cell, which operates at 1 x 103 to 3.5 x 103 Pa inlet pressure. The
elevated operating pressure relative to the supply pressure provides a reserve to accommodate flow up tran-
sients to the fuel cell.
4.2.1.2 Regeneration Process - The refrigeration process is represented in a block diagram shown in Figure
4-6. This process heats clean product landfill gas from the production process and regenerates the dryer and
hydrocarbon adsorption beds in the reverse flow direction during their regeneration cycle and destructs the
spent regenerant gas in an enclosed flare. An electric heater is used to heat the recycled clean landfill gas to
288 ฐC. This heated, regeneration gas is used first to regenerate the hydrocarbon adsorption bed. Second, the
dryer bed is regenerated. Third, the regeneration gas heater is bypassed and the dryer bed is cooled down with
cold regeneration gas. Lastly, the hydrocarbon adsorption bed is cooled down. Each heating and cooling
period lasts about two hours for a total regeneration cycle of eight hours. During transition from adsorption to
regeneration modes the regeneration gas is bypassed around the beds. At all times the regeneration gas flows
to the enclosed flare ensuring continuous operation of the flare and continuous thermal destruction of the con-
taminants and regeneration gas prior to atmospheric dispersion.
4.2.1.3 Refrigeration Process - The refrigeration process shown in Figure 4-7 uses R-22 refrigerant in the
cycle which provides refrigerated d-limonene coolant at a nominal 2ฐC to the first stage cooler and -28ฐC to
the second stage refrigeration cooler. The d-limonene refrigerant is accepted as an environmentally benign
organic extracted from orange peels and pressed pulp. The properties of d-limonene are given in Appendix F.
The refrigeration process incorporates a double-stage hermetically-sealed compressor and plate-type evapo-
rator. The refrigeration cycle operates to maintain the d-limonene coolant temperature setting at its discharge
from the evaporator. The compressor is driven by a 7.5 kW motor drive and operates continuously to recircu-
late R-22 refrigerant in the refrigeration process. The process operates with a claimed greater than 99 percent
reliability based on past operating experience. Both refrigerant R-22 and d-limonene coolant are completely
recycled and are not purged or vented from the process.
7. "Recovery of VOC's Using Activated Carbon"; Graham, James R. and Ramaratnam, Mukuno; Chemical
Engineering, February 1993
-37-
-------
International Fuel Cells
FCR-13524E
FROM LFG PRODUCTION PROCESS
+ 10'C
1
REGENERATION GAS HEATER
, 288ฐC (HOT REGENERATION)
+10ฐC (COOL DOWN)
DRYER ADSORPTION BEDS (2)
BEDHEATSUPT0232"C
HYDROCARBON ADSORPTION BEDS (2)
BED HEATS UP TO 204ฐC
VAPOR/LIQUID SEPARATOR
V f
NATURAL
GAS SUPPLY
(PILOT ONLY)
ENCLOSED FLARE
FLARE EXHAUST
Figure 4-6. Regeneration Process
13524-14
952104
-38-
-------
International Fuel Cells
FCR-13524E
COMPRESSOR
FINFAN CONDENSER
LIQUID RECEIVER
FILTERyDRYER
ฃ
d-LIMONENE FROM
1ST& 2ND STAGE
REFRIGERATION
COOLERS
EVAPORATOR
I
^
te
'v TO d-LIMONENE
\ SURGE TANK
*H 1ST & 2ND STAGE
/ REFRIGERATION
I COOLERS
^-_
Figure 4-7. Refrigeration Process Unit
-39-
-------
International Fuel Cells
FCR-13524E
4.3 PC25 Power Plant Design Modifications
4.3.1 Introduction and Background
In the Phase I Conceptual Study a preliminary list of modifications to the fuel cell were identified to achieve
80 kW, 140 kW, 175 kW and full rated 200 kW power. These preliminary studies were based on a nominal
gas composition of 50% methane and 50% carbon dioxide with a 4.45 kcal per standard liter heating value.
At the time of the Phase I Conceptual Study, natural gas blending with the landfill gas was considered as a
means to maintain the gas heating value to the fuel cell at 4.45 kcal per standard liter when the heating value
of the landfill gas was less than 4.45 kcal per standard liter.
4.3.2 Phase II Summary
The results of the detailed Phase II engineering study to identify modifications required for the fuel cell power
plant to operate on landfill gas are summarized in Table 4-5. The detailed Phase II study is based upon Pen-
rose site landfill gas analyses (43.9% mediane, plus 40.1% carbon dioxide, plus 15.6% nitrogen, plus 0.4%
oxygen) with a heating value of 3.91 kcal per standard liter. Natural gas blending was dropped as an option,
since the use of natural gas would lessen the value of the landfill-gas-to-energy demonstration to demonstrate
ability to operate on a wide range of landfill gas, including low heating value. Of the four power levels consid-
ered, the 140 kW nominal output was selected as the appropriate level for the demonstration, since the identi-
fied power plant modifications could be readily installed in the field after the initial fuel cell startup and
checkout on natural gas was completed.
Table 4-5. Modifications to PC25 A for Operation at 140 kW in Landfill Gas Demonstration
CHANGE TYPE
I. OPERATE TO 140 KW ON LANDFILL GAS
1 . Modify Control Software
2. Cathode Exit Orifice
3. Recycle Orifice
4. Inlet Fuel Controls
II. IMPROVE HALIDE TOLERANCE
5. Halide Guard Bed (Optional)
III. STARTUP
6. Stan Burners
DESCRIPTION OF CHANGE
Modify reactam flow schedules for landfill gas,
steam, burner air.
* Modify landfill gas flow transducer calibration
Modify fuel properties for landfill gas
Review process parameter event & shutdown lim-
its.
Review fuel control algorythms.
Reduce cathode exit orifice (FO 1 20) diameter to
balance stack cross pressure.
Enlarge recycle orifice (FO 3 10) to increase flow.
Install larger fuel control valve (FCV 012).
Install low pressure drop fuel venturi (FE012).
Add separate halide guard bed (catalyst, heaters,
thermal switch) within existing fuel processor.
Replumb start burners to natural gas supply. Add
solenoid valve, CV040, to reformer start burner
gas supply line.
The list of modifications in Table 4-5 is based on computer process simulations utilizing the measured landfill
gas composition available at the Penrose site. These simulations were used to verify that no other issues ex-
isted, and to provide a basis for the software revisions which are required as part of the changes. The detail
studies identified one potential issue. System simulations using the measured landfill gas composition indi-
-40-
-------
International Fuel Cells FCR-13524E
cated that pumping requirements for the steam ejector would exceed the ejector capability at 140 kW on land-
fill gas. Additional component tests were conducted using steam and air, instead of nitrogen, as originally
done. These more accurate tests indicate that the standard natural gas power plant ejector can provide 140
kW of process gas without modification. A total of seven modifications were identified for operation of the
PC25 A natural gas power plant in the landfill demonstration in the Phase I study, and these were verified
in Phase II. These modifications consist of four changes to achieve the 140 kW output power level, one
change to provide enhanced halide tolerance, one to provide enhanced tolerance to ammonia, and one change
to permit startup on natural gas after the power plant process inlet has been connected to landfill gas. These
changes are described in more detail in the subsequent sections.
4.3.2.1 Modify Control Software Control software changes permit the power plant to operate when the
input process fuel properties change from natural gas with a nominal heating value of 8.89 kcal per standard
liter to landfill gas with a heating value of only 3.91 kcal per standard liter. The input fuel properties and
reactant schedules are changed to compensate for the difference in heating value and chemical content of the
new landfill gas fuel. The changes in fuel properties and reactant schedules ensure that the power plant con-
troller will provide the appropriate level of process fuel, steam, and burner air for the power plant. The fuel
calibration curve for the landfill gas fuel flow transmitter is updated to ensure that the relationship between
volumetric fuel flow and energy content is consistent with the properties of the landfill gas provided to the
power plant. The fuel control software modifications assure accurate fuel flow control while operating on
landfill gas.
4.3.2.2 Cathode Exit Orifice - The cathode exit flow orifice, FO120, is modified to maintain an acceptable
cell cross pressure. This is accomplished by reducing the orifice diameter to increase the total pressure on the
cathode side of the stack, and balance the increased anode side pressure caused by the increased volumetric
fuel flow rate of dilute landfill gas.
4.3.2.3 Recycle Orifice - The recycle flow orifice, FO310, is enlarged to increase recycle gas flow to the fuel
processor.
4.3.2.4 Inlet Fuel Controls - The inlet fuel controls will be modified to reduce pressure drop. This is accom-
plished by installing a larger fuel flow control valve, FCVOI2, and installing anew fuel venturi, FE012, with
a lower pressure drop characteristic. These changes facilitate the higher inlet volumetric fuel flow of landfill
gas required to provide adequate heating value to the fuel cell.
4.3.2.5 Halide Guard Bed (Optional) - The Phase I study determined that a pretreatment system would
have to be designed to achieve total halide removal down to 0.15 ppm by volume to ensure a one year fuel
processor life. This compares favorably to the maximum of 0.032 ppmV total halides which was measured
during the Field Test (see Section 5.3.3). Therefore addition of a separate halide guard was not required. The
total halide content exiting the GPU was monitored during the field test to insure that the halide concentration
remained at safe levels. (See Section 6)
The halogen guard bed would be installed as a separate spool piece within the existing fuel processor section
of the PC25 A fuel cell powerplant. This guard bed would consist of 79 liters of a sodium oxide based adsor-
bent, heaters to maintain operating temperature of approximately 260ฐ C, and a local thermal switch to control
the heaters. The guard bed will remove halides including fluorides, chlorides and bromides which are hydro-
genaied in the existing PC25 A fuel processor.
4.3.2.6 Startup - The demonstrator PC25 A power plant will continue to use natural gas for the start burners
which preheat the stack coolant and the reformer to their normal operating temperatures during initial startup.
When the power plant process feed is changed to landfill gas, this will require replumbing the start burner gas
supplies to natural gas. At this time a solenoid valve, CV040, will be added to the reformer start burner line as
required by code.
-41-
-------
International Fuel Cells
FCR-13524E
4.4 Site Specific Process Design
4.4.1 Overall System and Site Description
A simplified description of the overall landfill-gas-to-energy demonstration is shown in Figure 4-8. The
demonstration consists of the landfill gas wells and collection system provided by Pacific Energy at their
Penrose Site, a modular gas pretreatment system, (described in greater detail in Section 3.1), a 200 kW PC25
A natural gas fuel cell power plant manufactured by ONSI Corporation modified for landfill gas operation,
a cooling module, and an interconnection to the grid. Landfill gas from four separate landfills is collected
at the site and compressed to 90 psig at the existing Pacific Energy facility before it is conveyed to the gas
pretreatment unit where water vapor and contaminants are removed to very low levels for the fuel cell. The
clean landfill gas is then converted to ac power for sale to the electric utility. Cogeneration heat generated
by the fuel cell power plant will be rejected by an air cooling module for the demonstration.
Utility power
lines
Gas wells and
and
collection system
Gas pre-
treatment
system
PC25A
fuel cell
power
plant
AC power
to grid
Landfill
Heat rejection
air cooling
module
Natural gas
Figure 4-8. LFC Fuel Cell Demonstration Program
A simplified process description of the landfill gas pretreatment and fuel cell system is provided in Figure
4-9. The demonstration starts with the raw landfill gas at a flow rate of 2260 standard liters per minute. The
landfill gas consists of 43.9% methane, 40.1% carbon dioxide, 15.6% nitrogen, with 0.4% oxygen on a dry
gas basis. The gas also contains 130 to 475 ppm by volume non-methane hydrocarbons, which includes con-
taminants consisting of 45 to 65 ppmV total halides (measured as chlorides), about 100 ppmV hydrogen sul-
fide, and an additional 11 ppmV of organic sulfur compounds (measured as H2S). The gas is saturated with
water at 6.2 x 105 Pa at the inlet to the gas pretreatment unit.
The first stage in the gas pretreatment process is a carbon bed which removes virtually all of the hydrogen
sulfide. This bed is not regenerated on site, but the carbon can be removed and regenerated off site if desired.
The gas is next regulated down from 6.2 x 105 to 1.5 x 105 Pa, before being cooled to approximately 2ฐC
in the first stage refrigeration condenser. The condenser stage reduces most of the water and some hydrocar-
bons, which are removed from the system as a condensate and relumed to the existing Penrose site condensate
treatment system. The next step is a regenerate adsorption bed which removes the water vapor to a dew point
of minus 50ฐC and also removes additional sulfur and halides. The dry gas is then passed over a second stage
cooler where the gas temperature is reduced to -28 ฐC before going through a regenerable activated carbon
bed for final removal of trace hydrocarbons, sulfur, and halides. The final step is fine pore filtration to remove
any particulates or dusting which may come from the regenerable adsorbent beds. The clean dry gas is regu-
lated down to about 3.5 x 103 Pa pressure. The resulting gas to the fuel cell is approximately 1560 standard
liters per minute at 3.5 x 103 Pa pressure with major contaminants reduced to less than 0.05 ppm by volume
total halides and total sulfur.
-42-
-------
RAW LFG INPUT CONDITIONS
2260 SUM
MAJOR CH4/CO2/N2
0.4% O2
130 TO 475 PPMV NON-METHANE
HYDROCARBONS
CONTAMINANTS
45 TO 65 PPMV
TOTAL HALIDES (AS CL)
- 100 PPMV H2S
- 11 PPMV ORGANIC SULFUR
(AS HZS)
WET
CLEAN LFG TO FUEL CELL
1560SL/M
MAJOR CH4/CO2/N2
0.4 O2
< 0.032 PPMV TOTAL
HALIDES (AS CL)
< 0.047 PPMV TOTAL
SULFUR
DRY (-50'C DEWPOINT)
<0.5 MG/DSCM
PARTICULATES
1.5x105Pป
COMPENSATE
TREATMENT
SYSTEM
NON-
REGENERABLE
HjS REMOVAL
*J* "i
COC
COND
CONDENSATE
RETURN TO
PENROSE
>LEfl
ENSER
.1
^
REGENERABLE
DRYER
ป
j
LOW
TEMPERATURE
COOLER
t
I
REGENERABLE
ACTIVATED
CARBON
BED
t
1
3.5 * 103 Pป
PARTICIPATE
FILTER
ง.
o
a
O)
CLEAN EXHAUST
COj4 HjO
t Nz + O,
140 kW
UTILITY
GRADE
POWER
FUNCTION
REMOVES
HjS
REDUCES WATER &
HYDROCARBONS
REMOVES
WATER,
SOME
SULFUR
AND
MAUDES
REMOVES TRACE
HYDROCARBONS
SULFUR 8
HALIDES
700 SUM
REMOVE
PARTICULATES
REGEN
GAS
HEATER
HEATS GAS TO
REGENERATE
BEDS
GAS PRETREATMENT SECTION
JL
POWER MODULE
FUEL
PROCESSER
- REMOVES
02
- REMOVES
RESIDUAL
SULFUR.
HAUDES
- CONVERTS
FUEL TO
HYDROGEN
CELL
STACK
- CONVERT!
AIR AND
HYDR-
OGENTO
DC POWER
AND
THERMAL
ENERGY
INVERTER
- CONVERTS
DC TO AC
- PROVIDES
SAFE
ELECTR-
ICAL
INTER-
CONNECT
COOl-
MODULE
REJECTS
WASTE
HEAT
NATURAL
GAS
FOB
START
BURNERS
AIR
FUEL CELL SECTION
Figure 4-9. Demonstration Project Processes
13524-16 <"
952404
to
-------
International Fuel Cells FCR-13524E
The dryer and carbon beds are regenerated using clean dry landfill gas from the exit of the pretreatment sys-
tem. The regeneration gas is heated to 288ฐC in an electric heater, and passed counter-current through the
regenerable dryer and activated carbon beds to remove the water and contaminants. The contaminants are
destroyed in a low NOx flare.
The fuel cell section of the demonstration consists of three major subsections: the fuel processor; the cell
stack; and the inverter. The fuel processor removes oxygen and any residual sulfur in the landfill gas and
converts the landfill gas to a hydrogen rich fuel. The next stage in the fuel cell power plant is the cell stack.
The stack converts air and the hydrogen rich fuel from the fuel processor to make dc power and thermal ener-
gy. DC power is sent to the inverter which converts the dc power to 60 cycle ac at 480 volts and provides
safe, electrical interconnect to the grid. All heat from the power plant during the demonstration will be re-
jected to the air by the cooling module.
The fuel cell power plant emits a clean exhaust stream consisting primarily of carbon dioxide, water vapor,
nitrogen and oxygen. Typical secondary emissions for the natural gas powered PC25 A fuel cell is about 0.5
ppmV NOx, 1.1 ppmV carbon monoxide, and 0.03 ppmV non-methane hydrocarbons, all measured at 15%
oxygen on a dry gas basis. During operation on landfill gas in Phase HI the fuel cell exhaust emissions were
measured as follows: NOX = 0.12 ppmV; and carbon monoxide = 0.77 ppmV. These results are discussed
in more detail in Section 6.3.3.
For a more detailed description of the gas pretreatment section process, refer to Section 4.2.
4.5 Site Specific Engineering Design
4.5.1 Site Location
The landfill gas to energy site is located at Pacific Energy's Penrose Power Station in Sun Valley, CA. Selec-
tion of this site was based upon site selection criteria which were developed and discussed in Section 4.1.
The location of the selected site (labeled ฉ) at the Penrose Power Station facility is shown in Figure 4-10.
4.5.2 Site Arrangement
The site arrangement for the landfill gas to energy demonstration is shown in Figure 4-11. The demonstration
siie is completely enclosed by a chain-link fence. The gas pretreatment skid and refrigeration unit are
installed in the middle of this demonstration site area in the lined zone. This area represents the limit of the
zone which is classified Class I, Division 2 for electrical equipment by the National Electrical Code due to
me presence of the gas pretreatment system. Lying outside this area to the north is the control panel for the
Gas Pretreatment Unit control panel which is located inside an existing utility building. This building has
also been outfitted with phone communications. To the right of the existing utility building is the flare for
the gas pretreatment system. This flare functions to destroy the contaminants removed from the landfill gas
while emitting low levels of secondary emissions.
The fuel cell power plant and cooling module are located directly due south of the gas pretreatment unit as
shown in Figure 4-11. The fuel cell power plant installation is completely standard for an ONSI natural gas
fuel cell power plant. The fuel cell pad is made of reinforced concrete approximately 335 cm wide by 762
cm long, while the cooling module pad is approximately 244 cm by 366 cm.
-44-
-------
International Fuel Cells
FCR-13524E
Figure 4-10. Fuel Cell Site Options: Site 2 Selected for Demonstration
(Courtesy of Pacific Energy) <13047-09)
-45-
-------
International Fuel Cells
FCR-13524E
CONTROL PANEL
UTILITY QUILOING
GAS
PRETREATMENT SKID
REFRIGERATION UNIT
DOST. NATURAL CAS METER
FUEL CELL SKID
FENCE GATE
COOLING MODULE
. j^
u. _. __ _ _.-_.,_..,
-1
GAfE TO PLANT
i
ll
13524-18
9531 CM
Figure 4-U. Site Layout (13047-09)
-46-
-------
International Fuel Cells
FCR-13524E
4.5.3 Site Design Details
The details of the EPA landfill gas site design are described in seven site blueprints shown in Table 4-6, which
cover site plan and details, structural specifications, foundation plan and schedule, structural details, mechan-
ical plan, mechanical piping plan, and P&ID drawing.
Table 4-6. Summary of Detail Site Design for EPA Landfill Gas Demonstration
PRINTS
C-l
S-l
S-2
S-3
M-l
M-2
PI-1
DESCRIPTION
Site Plan & Details
Structural Specifications
Foundation Plan & Schedule
Structural Details
Mechanical Plan
Mechanical Piping Plan
P&ID Drawing
-47-
-------
International Fuel Cells FCR-1352E
5.0 GAS PRETREATMENT UNIT VERIFICATION TEST
This section describes the verification testing to confirm the performance of the gas pretreatment unit (GPU)
prior to connection to the fuel cell power plant for the field demonstration of recovery of energy from landfill
gas in Phase III. Major objectives of the verification test are: (1) process performance verification (demon-
strate that the GPU meets contaminant removal requirements for the fuel cell of less than 3 ppm by volume
total sulfur and less than 3 ppm by volume total halogens); (2) reliability demonstration of up to 500 hours
total and at least 200 hours continuous operation; and (3) emissions testing of the GPU flare required by South
Coast Air Quality Management District.
The specific test plan to achieve these objectives is discussed in Section 5.1 - Landfill Gas Pretreatment Mod-
ule Test Plan. Section 5.2 discusses permitting requirements for the South Coast Air Quality Management
District (SCAQMD) and L.A. City. All GPU test results including factory test, field checkouts, and the Phase
II field test and SCAQMD tests, are discussed in Section 5.3 - Test Results.
5.1 Landfill Gas Pretreatment Module Test Plan
IFC developed a Test Plan (Appendix B) to ensure that all GPU objectives established in Phase I would be
met.
The testing was divided into three parts: (1) Factory test at IFC in South Windsor, Connecticut, on nitrogen
to verify the thermal, mechanical, and electrical operability of the GPU; (2) Site Checkout Test at the Penrose
site in Sun Valley, California, on landfill gas to determine if adjustments were required prior to conducting
the Phase II EPA field test; and (3) Phase II EPA Field Test to document contaminant removal performance
and flare performance.
The Factory Test and Site Checkout Test are described in Appendix B, Section 2.0. These tests are designed
to identify any mechanical or process issues early so they can be addressed and corrected prior to initiating
the Phase II Field Test.
The Phase II EPA Field Test is described in Appendix B, Sections 3 and 4. Section 3 describes the Test Plan
for the gas pretreatment unit process performance measurements to ensure that the contaminant removal of
sulfur, halogens, and particulates from the landfill gas meet the requirements for the fuel cell power plant in
the Phase III demonstration test. Section 4 describes the test for the South Coast Air Quality Management
District permit requirements which include the emissions characteristics of the GPU flare. The complete test
protocol for the Field Test is summarized in Table 5-1.
The results of all GPU testing are described in Section 5.3 Test Results.
-48-
-------
Table 5-1. Test Protocol for Phase II BPA Field Test
TME KET
PEWOO WERATH8 CM
Mr (HP) BED CONTOIS
PRE-Tฃ3T*AWnฃ/MALYS(6-
- OTW. FTU CWCXOUT AKALY9B -
"PTUTE4THG
1 OMMS3C A IKป
ซป09W A ซซ
OHO-ISOO A .R-tt-
ISaO-tOO * ปf>-l?
NB9-2CQ B IFG
2 OOXMUOO A LFC
ttOmn B Lซ
O900.)i00 B IK)
1500-1600 B LFG
WOO-2400 A LFG
1 ODCfrOMO B LFC
MOO-MOO A LFO
0900-1500 A LFO
1MMM) A LFO
SHTU-VHX LOCATION
FffwaTYPt
SMflE AKW.VHS ECMWNT
OUTLET 0ซ l*AeWWOT9-f1U TO FUa CEIL]
SPECUL
8CAOM)
TOT* THSWO
MFUfl H2S INDMXW. R-1P TOTAl
OW.Y ONLY HAUX& OW.Y MRtKUATE TABLE B|
ON9TE ON9IE ON9TE MOTE OFFHTC OfFSTTE
G024tM ' OOKO ซo
(oninMiiiMp
CMMM^M)
1HR IHR 1HR 1HR V 24HR
_ _ _ _ _ _
TE6 - Ytt ~ -
YEป WS
YES
ซ
1t8 YE6 YEซ
WKWttalmMunvto*** ซS
YES YES YES
YH TCS YES ซ8
>ES tES YES >E5
YES YEJ
YES YES YEB YES
MM 8V1J3 3VXC MOO SVW 8V300
BOTM aAปflEV*eH*VE MUCH SWWEIO* HALE RTW06
1 J SOJ *ปiซt v**4 tanซปrt 802 <**r
ป|OMO> !* dซm*orซ(*l fbctren ufM Mitto
1| BUi.lCMiMlnc n>inปซnMซl*n Do
HET GAS MEASUR9*XII . (UWCHJ. TO PW)
TOTN.C WLA1U flEMENTA]. ffEOU.
KDMWAL PWOWTY. aUCOMfac SOWMD
HJlfUR Ht.fl-U- 9Urfซ* TE8TW
B-1? COMPOUHW POUlfTAKT* SLOXWES
Ot.Y (TABLE*) {TABLE q FWJfO. (TABUW (TAaEB)
OKSITE OM9TE OFF9TE OFfart OFF9TTE OfFSHE
fM kig k<( Ml k^l tag
[*<*ป GOFPO OCBtt. KPOL MS
SOura 60TO
lMi|
1HR IHR tT W -8* }4HR
YES YE3 YES ซS
_ _ _ _
YE9 YES YES ซ9
YES
YEB YES Yf6 rtซ
VEfi
tES YES VES YES
8V708 6Y506 SWOB SV206 SWM SVW
V4NCH SWAOaOR IWli FITTOG
5JOQM3-.VM dnmซi>fn(*ylRua mektwnr
(.) OCH>. |M ibatutttnftt} 1 !ซ< fcniittM *ซllt*tซ
7 )KPlfc H|K pnitn I** dmMbpvtir
t )JM9: *mc ibwpb* indntem
CONOEKSAIE5
TOIM.OAOUUDCS
(BOAM)
VESSEL VEsea
1 2
OFFOTE
tซM f*M
34 HR 24 HR
_ _
-
YES
YES
ซ6
YES
YEJ
YES
svi7i svin
WWฃ
FLWEIESTIW
SfEOAL SCAOWTESIWG'
FUIC FLAK
MฃT EXMAUS1 AHBENT
(TABLE B) (TABtfE) (TABLEE)
OH=6flE OfFCtTE OFfMTE
tng big tป9
MKR 24HR 24HR
- - VES
VWLT1 UU.1I WX.T1
YES YES YES
YES YES YES
ฃV1ซ STMffiWe -
VWฃ SAMPIEPCATS
COUMENTB
MULYSeCOMPMYlOCATOtt
bulhfiEIMOO
ANALVaBEOUHCXT USED
TMNqFORPRSJUflESULTS
(ep^BHitotft*)
IRC CUBRATUN CHECKOUT
OPnOHALTROTESTKOTa
cowmi GAiHwiE PTU
fUe-QWUEKOE TEETMO,
OUaEHGETESTOFBGOA
UฃMGAPntOX.ttFnปOF
B-12-MOEDTOlfa
N ORMM. TEST Of SEOONO flEO
|MULT1-My|ptซtซiipta
MปndปM)
HonMALTESTOFFRSTBED
WTERATlฃAST24HtHJR
USE aNCECHMlfMOE TEST
n
ป
p
(*) R-12 = Dichloradllluoromdhane
-------
International Fuel Cells
FCR-1352E
5.2 Permitting
Obtaining local permits is an essential part of the demonstration project. For the landfill gas demonstration,
two major permits were required: South Coast Air Quality Management District for emissions; and the City
of Los Angeles Building and Safety Department for local building and safety codes.
5.2.1 South Coast Air Quality Management District Permit
The gas pretreatment unit flare requires a SCAQMD permit to construct and operate. A copy of the permit
to construct and operate is given in Appendix B, Attachment D.
The permit to construct and operate is based on 18 permit conditions. The key technical condition is given
in Condition No. 11 which requires that "a temperature of not less than 760ฐC as measured by the temperature
indicator shall be maintained in the flare stack." The second critical condition is No. 16 which requires that
performance tests be conducted and the results forwarded to the South Coast Air Quality Management Dis-
trict. The required SCAQMD tests were included as part of the test plan protocol discussed in Section 4.1.
These tests were conducted as part of the GPU qualification test conducted on October 19, 20 and 21,1993.
5.2.2 L.A. City Permits
The L.A. City Department of Building and Safety administers local codes to ensure that minimum safety
standards are met for mechanical and electrical equipment. This process includes submittal of site construc-
tion drawings for plan checks plus inspection of electrical and mechanical aspects of the gas pretreatrnem
unit system by field inspectors and/or the L.A. City Test Labs. For the gas pretreatment unit separate permits
were required for the pretreatment skid, the water chiller (refrigeration unit), and the flare. The pretreatment
skid and the flare have been approved. The water chiller has passed field inspection and all issues have been
settled with the L.A. City Test Labs. A brief chronological summary of the activities associated with obtain-
ing the L.A. City permits over a period of 17 months is summarized in Table 5-2.
Table 5-2. Permit Activities for EPA Gas Pretreatment
September 14, 1992
October 22, 1992
Januarys, 1993
February 19, 1993
March 19. 1993
April 20, 1993
June8. 1993
June 11. 1993
August 2, 1993
February 24, 1994
Submitted applications for L.A. City Building and Safety Plan Checks.
Received plan correction sheet from L.A. requesting additional information.
L.A. City Building and Safety Permits are approved.
Installation Contract awarded to Unit Construction.
Additional Electrical permit required
Pretreatment Skid, Water Chiller and Flare required an additional Special
Equipment Permit. Pretreatment Skid and Flare was approved. The Water
Chiller required approval from the L.A. City Test Labs, because it was a
listed piece of equipment.
Application for the Water Chiller was submitted to the L.A. City Test Labs.
Application for a variance to operate Water Chiller was submitted. Variance
was approved on June 24, 1993.
Field inspection was made by L.A. City Test Labs on the Water Chiller, a
L.A. City Test Lab report was issued on August 5, 1993.
All of the issues have been settled with L.A. City Test Labs.
-50-
-------
International Fuel Cells FCR-1352E
5.3 Test Results
This section describes the results of the tests described in Section 5.1, Landfill Gas Pretreatment Module Test
and Quality Assurance Plan. These test results are reported in three sections: (1) Factory test on nitrogen
to verify the thermal, mechanical and electrical operability of the gas pretreatment unit; (2) Site Checkout
Tests on landfill gas to determine if adjustments are required prior to conducting Phase II EPA Field Test;
and (3) the Phase II EPA Field Test to document contaminant removal performance and document flare per-
formance.
5.3.1 Factory Test Results
The Factory Test on nitrogen was conducted at International Fuel Cells' facility in South Windsor, CT. The
test demonstrated that the Gas Pretreatment Unit (GPU) can be operated and controlled at its designed temper-
ature, and verified the pressure drop in the GPU at design flow is approximately 4.1 x 104 Pa with nitrogen
gas. This compares favorably with design value of 3.4 x 104 Pa psi pressure drop with landfill gas. Based
upon this successful factory test, the GPU was shipped to the Penrose site in Sun Valley, California. The test
results are described in more detail in Appendix B, Attachment C.
5.3.2 Site Checkout Test Results
The initial Site Checkout Test was performed at the Penrose site between April 25 and May 2, 1993, after
repairs were made to correct minor damage which occurred during shipping. The GPU was operated briefly
on landfill gas to assess operability and gas pretreatment performance. An analysis of gas samples taken dur-
ing operation indicated that the hydrogen sulfide (H2S) present in the landfill gas was being converted to car-
bonyl sulfide (COS) during processing in the GPU.
Laboratory testing confirmed that the source of carbonyl sulfide is reaction of hydrogen sulfide with carbon
dioxide in the landfill gas over dry (regenerated) alumina desiccant:[H2S + CO2 ^ COS + H2O]. The
laboratory testing confirmed that mass extraction of water vapor by the dry regenerated alumina drives the
reaction to almost 100 percent conversion to COS. (See Appendix F).
The high concentration of carbonyl sulfide would cause the gas to the fuel cell to exceed the 3 ppm V specifica-
tion, since carbonyl sulfide removal from the landfill gas by the downstream activated carbon bed is limited.
1FC therefore elected to eliminate the carbonyl sulfide formation by removing the hydrogen sulfide from the
raw landfill gas upstream of the first stage condenser. An activated carbon impregnated with potassium hy-
droxide was selected to remove the H2S. Testing of the impregnated carbon material on landfill gas has
shown that it removes all H^S, plus some organics, and halogens from landfill gas. The H2S removal was
also confirmed in laboratory tests at IFC and the manufacturer. The IFC and manufacturer test data are sum-
marized in Appendix C. Based upon these tests, the GPU was modified by the addition of a single H2S remov-
al bed, upstream of the first stage condenser.
The second Site Checkout Test was conducted between September 6 and September 8,1993. This checkout
confirmed complete H2S removal in the new bed, validated that the gas pretreatment processes were function-
ing as expected, and confirmed that the GPU was ready for verification testing. Gas sampling was taken dur-
ing 3 days of operation. Operation during that time was not continuous, due to a problem with the refrigera-
tion unit, which caused several shutdown/restarts. The problem was determined to be loss of Freon due to
a leak, probably caused during shipping of the unit. The leak was repaired and the unit recharged, which
restored normal operation. The unit accumulated approximately 36 hours of operation, during which 21 bag
samples were taken at strategic sampling points in the process.
The results of the H2S analysis, as well as those of other sulfur compounds, are shown in Table 5-3 at a point
in time approximately 18 hours into the checkout test and at eight different sample locations in the GPU.
As predicted by lab tests, the results show that all the H2S was removed to below detectable levels in the new
H2S removal bed. No COS was formed in the alumina dryer bed (C), and overall, there was no detectable
sulfur at the exit of the system (E). One species, dimethyl sulfide, which appears to be absorbed on the dryer
-51-
-------
International Fuel Cells
FCR-1352E
bed (C) is removed during regeneration (G) indicating that sulfur is not building up in the dryer bed. Based
on these results, it was decided to proceed into the GPU Field Testing portion of die program.
Table 5-3. Gas Pretreatment Unit Sulfur Removal Performance
-18 Hours into September 1993 Site Checkout Test
SPECIES
Hydrogen
Sulfidc
Carbonyl
Sulfidc
Methyl
Mcrcapi&n
El Ml
Mcrcaptan
Dimctlul
Sulfidc
Carbon
Disul fide
Dimclh>l
Disulfidc
Exit Gas Concentration (PPM) at Sample Port Locations
LFG INTO
UNIT
80.3
0.16
3.26
0.51
6.25
0.06
ND(1)
A
HjS
REMOVAL
ND{2)
ND(2)
ND(2)
ND(2)
9.07
0.09
0.09
B
COOLER
CONDENS-
ER
ND(2)
0.07
ND(2)
ND(2)
8.25
0.09
0.1
C
REG EVER-
ABLE DRYER
ND(2)
0.08
ND(2)
ND(2)
ND(2)
0.03
ND(1)
D
LOW
TEMPERA-
TORE
COOLER
ND(2)
ND(2)
ND(2)
ND(2)
ND(2)
0.04
ND(1)
E
ACTIVATED
CARBON BED
ND(2)
ND(2)
ND(2)
ND(2)
ND(2)
ND(1)
ND(1)
F
REGENERATION GAS
FROM ACTIVATED
CARBON BED
ND(2)
0.59
ND(2)
ND(2)
ND(2)
0.60
ND(1)
G
REGENERATION GAS
FROM DRYER BED
ND{2)
ND(2)
0.31
ND(2)
34.9
ND(1)
0.67
ND = None Detected
n) - 0.02 ppm Detection Limit
(2) - 0.04 ppm Detection Limit
5.33 Phase II, EPA Field Test
The Phase II Field Test of the gas pretreatment unit (GPU) was successfully completed during October 20,
21, and 22, 1993 according to the Test Plan (Appendix B) discussed in Section 5.1. The results from this
testing are provided in a report by TRC Environmental, Inc., in Appendix D. These results show that the GPU
operated well within process specifications for sulfur and halide removal. Total particulates at the exit were
less than detectable. Emissions from the flare are consistent with SCAQMD requirements. Subsequent to
the Field Tests in October, the operational reliability of the GPU was successfully demonstrated and the unit
was voluntarily shut down on March 3,1994 after completing 216 hours of continuous operation and a total
of 616 hours since first startup. Based upon these results, the GPU was declared ready for operation with
the fuel cell power plant in the Phase III Field Test Demonstration.
The Phase II Field Test was performed under the direction of IFC personnel with assistance from Pacific Ener-
gy site personnel. TRC Environmental, Inc. was responsible for all gas analyses. Testing was conducted over
a 3 day period. "Day 1" was devoted primarily to challenge testing of the unit's "A" or even numbered regen-
erable dryer and activated carbon beds by adding a light halogenated species to the landfill gas entering the
GPU. Dichlorodifluoromethane was selected as the challenge gas for this test. Raw inlet gas composition
and outlet gas composition were analyzed according to the requirements given in the test protocol in Table
5-1. "Day 2" was devoted to testing of the "B" or odd number regenerable beds with raw landfill gas without
the dichlorodifluoromethane addition and to testing the flare for the South Coast Air Quality Management
District. Additional gas analysis was performed on the raw inlet gas, outlet gas, and regeneration gas flare
inlet and exhaust as well as ambient air reference samples. The flare inlet and exit were monitored three times
during regeneration per the test protocol in Table 5-1. "Day 3" was devoted to retesting of the "A" beds on
raw landfill gas without dichlorodifluoromethane addition following at least a 24 hour period since the addi-
tion of dichlorodifluoromethane challenge gas on "Day 1." Raw inlet gas and outlet gas compositions were
analyzed. The "Day 3" testing was initiated immediately following the "Day 2" activities to expedite the
verification procedures.
-52-
-------
International Fuel Cells
FCR-1352E
The results of the Phase II Field Test of the GPU are in Figure 5-1 and Table 5-4. Figure 5-1 shows the sam-
pling locations, dichlorodifluoromethane injection location, plus inlet and outlet gas conditions on a simpli-
fied block diagram of the GPU including the new P^S removal bed at the inlet. The tests were all conducted
with an inlet landfill gas flow of 2260 standard liters per minute, and an outlet flow of 1560 standard liters
per minute of clean landfill gas, which is sufficient to generate 140 kW of continuous electric power using
the PC25 A fuel cell power plant. The output conditions in Table 5-4 show that the major contaminant remov-
al requirements have been met by a wide margin. Total halogens (measured as chloride) were less than or
equal to .032 ppmV which is over 90 times less than the 3 ppmV requirement. Total sulfur (measured as P^S)
was less than or equal to .047 ppmV which is over 60 times less than the requirement of less than 3 ppmV.
The overall results of the Phase II field testing of the gas pretreatment unit are summarized in Table 5-4. Total
halogens (as chloride) were reduced from an average inlet concentration of 60 ppmV (average of 6 tests) to
an outlet concentration ranging from non-detectable to 0.032 ppmV. The only species detected at the exit
was methylene chloride. Taking the highest total outlet halogen level (0.032 ppmV as chloride) divided by
the average inlet total halogens of 60 ppmV as chloride yields a removal efficiency of at least 99.95 percent
for halogens. During the dichlorodifluoromethane challenge test an inlet level of 7.4ppmV dichlorodifluoro-
methane was reduced to non-detectable (less than 0.002 ppmV) for a removal efficiency of greater than 99.97
percent. Total sulfur (measured as H2S) averaged 113 ppmV at the H2S adsorber inlet(average of 3 tests),
with outlet levels of non-detectable to 0.047 ppmV, for an overall removal efficiency of at least 99.96 percent.
The only sulfur species detected at the GPU exit was carbonyl sulfide. Particulates were measured at the GPU
exit on three occasions with all three test results being less than detectable. The third test was run at the lowest
detection limit of 0.5'Mg per dry standard cubic meter, indicating that the gas going to the fuel cell is virtually
paniculate free. Silanes and siloxanes were also measured several times with increasing sensitivity but were
at all times less than the detection limit at the inlet. The third test found no measurable silanes or siloxanes
at a detection limit of 0.076 milligrams per dry standard cubic meter. Consequently, no measurements were
taken at the GPU outlet. Likewise, phenol was measured and found to be less than the detection limit of 0.03
ppmV at the inlet.. In summary, the gas pretreatment unit is operating with an overall contaminant removal
efficiency of greater than 99.9 percent, and the gas available to the fuel cell power plant for the demonstration
easily meets all fuel cell requirements.
Condtnur 1
Outlet
dlchlorodl-
fluorome-
than
Monnorlng
citle
CondonMr 1
CondปnMlป
Sampl*
TO
FLARE
s
DRYER
BEDS
TO * ป
FLARE
CARBON
BEDB
1560
SUM
CLEAN
LFGTO
FUEL
CELL
REGENERATION
I CAS
I TOO SUM
REGEN
GAS
HEATER
OFF-LINE BED REGENERATION
13524-18
9S2104
Figure 5-1. Phase II Gas Pretreatment Unit Sample Locations. Shown with "A" Beds On Line,
and "B" Beds being Regenerated.
-53-
-------
International Fuel Cells
FCR-1352E
Table 5-4. Summary of Phase II Field Testing of Gas Pretreatment Unit
TEST
CATEGORY
Total Halogens (as chloride) W
dichlorodifluoromethane
Total Sulfur (as H2S)<3>
Particulates
Silanes, Siloxanes
Phenol
Non methane organics (as CH4)
INLET TO GPU
60ppmV
7.4 ppmV
llBppmV
< .076 MG
DSCM
< .03 ppmV
5700 ppmV
OUTLET OF GPU
ND to .032 ppmV
ND(2>
ND lo 0.047 ppmV <4>
<.05 MG
DSCM
13.8 ppmV
REMOVAL
EFFICIENCY <*>
99.95%
99.97%
99.96%
99.8%
REMARKS
ONLY species detected is me-
thylenc chloride
Challenge test with Freon- 1 2
added to GPU inlet
Major species detected is car-
bonyl sulfide
Below detection limit of 1 mg
Below detection limit at inlet
Below detection limit at inlet
.VOTES; < Total halogens as chloride = sum of each individual halogen compound delected times number of halogen atoms
(chlorine, fluorine) in that compound (e.g.. 1 ppmV ofFreon-12 (dichlorodifluoromeihane) = 4 ppmV as chloride)
t21 Non-detectable to .002 ppmV as the species
l}'Totai sulfur as H^S = sum of each individual sulfur compound limes the number of sulfur atoms in that compound
141 Non-detectable to 0.004 ppmV to O.OlOppmV.
(5 > Removal Efficiency =( } _ M,( } x loo
inlet
A good overall indicator of GPU cleanup performance is total nonmethane organics. Total nonmethane or-
ganics (as CH4) showed a reduction from 5700 ppmV at the inlet to 13.8 ppmV at the GPU outlet, for an
overall removal efficiency of 99.8 percent. In addition to hydrocarbons containing sulfur and halogen, the
nonmethane organics include nonhaloginated and nonsulfur species such as propane, butane, pentane, hex-
ane, benzene, toluene, xylene, acetone, ethyl acetate and heavier compounds such as d-limonene which were
all detected at low levels in the raw landfill gas. These compounds do not need to be removed from the landfill
gas for the fuel cell power plant, but the high removal efficiency of these compounds is a further indication
of the overall capability of the gas pretreatment unit.
The overall result of the Phase II Field Test of the GPU flare are summarized in Figure 5-2 and Table 5-5.
The data show that the flare achieves high destruction efficiencies even while operating on the regeneration
gas from the dryer bed "A" hot regeneration, which showed the highest contaminant levels to the flare. Sam-
ple destruction efficiencies during the dryer bed hot regeneration are consistently above 99 percent: dichloro-
methane at greater than 99.97 percent; tetrachloroethylene at greater than 99.85 percent, dimethyl sulfide at
greater than 99.2 percent; and total nonmethane organics at 99.2 percent. The flare efficiency test results are
discussed in greater detail in Appendix D.
-54-
-------
International Fuel Cells
FCR-1352E
Carbon Bed A
Hoc BeganwaUfln
Dryซr B*d A
Hot Hซgtnซratlon and
Cold Regeneration
X
TO
FLARE
DRYER
BED A
TO *-"
FLARE
hM
CARBON
BED A
1560
SUM
CLEAN
LFGTO
FUEL
CELL
REGENERATION
I GAS
700 SL/M
I
OFF-LINE BED REGENERATION
13524-19
952104
Figure 5-2. Landfill Gas Pretreatment Unit Sample Locations for GPU Flare Tests.
Shown with "B" Beds On Line, and "A" Beds being Regenerated.
Table 5-5. Landfill Gas Pretreatment Unit Field Test Results for GPU Flare
Data taken October 2 1 , 1 993 by TRC Environmental
ซ Gas Pretreatment Unit Inlet Flowrate 2260 SUM, Output 1 560 SL/M, Regeneration 700 SL/M
Requlrenซn, *ฃ?'
Time
Process
Requirement Activity
Sample
Location
AaMethane (ppmV)
B.Total Non-Methane Organic* (ppmV)
C.OxIdes of Nitrogen (ppmV) (5
D.Carbon Monoxide (ppmV) B
E.Total Particulales (Micrograms/M3) S)
From Half
Back Half (Organic)
Back Half (Inorganic)
F.HydrogenSulflde (ppmV)
G.C, through Cj Sulfur CPDS [Total AS HjS]
Carbonyl Sulfide (ppmV)
Methyl Mcrcaplan (ppmV)
Ethyl Mcrcaptan (ppmV)
Dimethyl Sulfidc (ppmV)
Carbon Disulfide (ppmV)
Dimcihyl Disulfide (ppmV)
H.Carbon Dioxide (ft)
Landfill gas pmreatment
unit
1000-1700
"B" Beds on-line, "A"
Beds on-rcgen,
GPU Inlet
472.000
5.700
NRQ
MR
NR
MR
NR
NR
106
117
0.16
2.79
0.44
6.57
<0.04
<0.04
o
GPU Out-
let
483.000
13.8
NR
NR
NR
NR
NR
NR
<0.004
0.017
0.017
<0.004
<0.004
<0.004
<0.002
<0.002
ra
Flare
1030-1130
Carbon Bed "A" Hot
Regeneration
Flare
Inlet
440.000
1.860
NR
NR
NR
NR
NR
NR
<.004
0.254
0.061
<0.004
<0.004
0.042
0.146
<0.002
13
Flare
Outlet
-------
International Fuel Cells
FCR-1352E
Table 5-5. Landfill Gas Pretreatment Unit Field Test Results for GPU Flare (Continued)
Requirement
Requirement
I.ToxIc Air Contaminants
Equipment
Tested
Time
Process
Activity
Sample
Location
Benzene (ppmV)
Chlorobenzene (ppmV)
1.2 Dichloroethane (ppmV)
Dichloromethane (ppmV) (Methylene chloride)
Teirachloroethylene ippmV) (teiraehloroelhene)
Teirachloromeihane (ppmV)
Toluene ippmV)
I.I.I Tnchloroe thane IppmV)
Tuchloroethylene (ppmV)
Tnchloromeihane (ppmV)
Vm>l Chloride ippmV)
X>lene (ppmV)
ADDITIONAL CONTAMINANTS
Dirhlorodifluoromethane (ppmV)
Ci* - 1.2- Dichloroethene (ppmV)
1.1 - Dichloroethane (ppmVi
Ethyl Benzene (ppmV)
Styrene(ppmV)
Acetone IppmV)
2 - Bulanone (ppmV)
Eihyl Acelate (ppmVl
hlh> 1 Buiyrate (ppmV)
Alphj Pmrnc tppmV)
d-Limonenc (ppmV)
Tcirah>droluran (ppmV)
J.OxxenCr)
K.Nilrogen I "* I
L.MotstuiT < * )
M.TnnperBture(!Fl
VFIrm kBiclSL/Ml
Landfill gas pretreatment
unit
1000-1700
"B" Beds on-line, "A"
Bed: on-regen,
GPU Inlet
1.7
1.4
<0.35
4.1
4.8
<0.23
47
<0.26
2.4
<0.29
1.4
28.2
0.26
5.8
2.8
12
1.1
15
3.7
10.8
8.4
18
18
2
(2
m
2260
GPU Out.
let
<0.002
<0.002
<0.002
<0.002
<0.002
<0.001
<0.002
-------
International Fuel Cells FCR-1352E
5.3.4 Conclusions from Phase U GPU Field Test
The Gas Pretreatment Unit (GPU) was successfully installed and tested at the landfill gas site in Los Angeles
(Sun Valley), California, and is ready for operation with the fuel cell power plant. The GPU functioned auton-
omously, purifying landfill gas to a level which is more than suitable for fuel cell use. In addition the GPU
flare has received permits from the South Coast Air Quality Management District, a district with very
strict air quality regulations. A summary of the performance of the GPU is as follows:
The GPU removed total sulfur in the landfill gas (measured as HiS) from 113 ppmV to < 0.047
ppmV. This is a removal efficiency of at least 99.96%, and is over 60 times better than the speci-
fied limit of 3 ppmV total sulfur at the exit.
The GPU removed total halogens (measured as chlorides) from 60 ppmV to < 0.032 ppmV.
This is a removal efficiency of at least 99.95%, and is over 90 times better than the specified limit
of 3 ppmV total halogens at die exit.
During a challenge test, the GPU removed 7.4 ppmV of dichlorodifluoromethane (added at the
inlet) to less than 0.002 ppmV at the exit. This represents a removal efficiency of greater than
99.97%.
The GPU flare safely disposed of the landfill gas contaminants removed by the GPU by achiev-
ing high destruction efficiencies above 99%. The flare was permitted by the South Coast Air
Quality Management District for operation in the Los Angeles area.
A total of 616 operating hours was logged on the GPU, including a 216 hour endurance run which
was voluntarily terminated.
-57-
-------
International Fuel Cells FCR-13524E
6.0 FUEL CELL DEMONSTRATION TEST
This section describes the results of the Phase III demonstration test activities beginning in January 1994 and
ending in June, 1995. During this third phase of the program, IFC developed a Test and Quality Assurance
Project Plan, completed all permitting activities for the fuel cell with the City of Los Angeles, installed and
checked out the fuel cell power plant at the site on natural gas, modified the fuel cell for operation on landfill
gas, and then connected the fuel cell to the gas pretreatment unit and operated the demonstration test, includ-
ing obtaining critical emissions and operating data. The demonstration operated at the existing Penrose Sta-
tion landfill gas energy recovery facility owned by Pacific Energy in Sun Valley, California. Internal combus-
tion engines presently generate up to 8.9 MW of electricity at this site with landfill gas from four separate
landfills. Electricity produced by the fuel cell was sold to the Los Angeles Department of Water and Power
electric utility grid.
6.1 Test and Quality Assurance Project Plan (QAPP)
The Test and Quality Assurance Project Plan (QAPP) for the Phase III Demonstration Test is given in Appen-
dix G. The QAPP describes the program objectives plus the test quality requirements, measurements, cal-
culations and quality audits required to assure that the proposed testing meets the EPA requirements. The
plan was written to meet the requirement of an EPA Category II quality assurance plan and a site-specific
test plan. The QAPP is designed to measure fuel cell and GPU performance while operating for an extended
period on landfill gas. Fuel cell performance includes efficiency, availability, operating and maintenance
costs, and emissions. GPU performance includes contaminant removal effectiveness of sulfur and halide spe-
cies which are deleterious to long term fuel cell performance. Results of these tests are described in Section
6.3.
6.2 Test Preparation
6.2.1 Permitting
All site construction and equipment installation activities require approval by local government authorities,
in this case the Los Angeles Building and Safety Department and the South Coast Air Quality Management
District for emissions. The Building and Safety and SCAQMD permitting for the GPU is described in Sec-
lion 5.2. The fuel cell has a blanket exemption from emissions from the SCAQMD, so no fuel cell emission
permitting was required.
Permitting activities with the L.A. Building and Safety Department were initiated in August 1993, to allow
adequate time, since this demonstration is the first-ever fuel cell power plant to be permitted for operation
within the jurisdiction of the City of Los Angeles. After working closely with the L.A. City Building and
Safety Department, a variance was granted in April 1994 to permit operation of the landfill gas fuel cell power
plant. The conditions of approval for the variance are shown in Appendix A, page A-3. Subsequent to the
variance, the normal plan checks of the site mechanical drawings were completed in May and June and the
on-site inspections of the mechanical installation were completed during July and August. Electrical site
inspections of the demonstration site equipment were completed during August with final approval in Sep-
tember. Final approval of the electrical drawings was delayed due to the added requirement to install load
shedding equipment to protect an existing grid interface transformer in Pacific Energy's Penrose plant. This
equipment was installed during September, with final approval coming in October 1995.
6.2.2 Site Preparation
The bulk of the site preparation for the fuel cell power plant and cooling module was completed in 1993, when
the GPU was installed on the site. Final site preparations, including repair of the fuel cell foundation drain,
and extension of the concrete mounting pad for the cooling module, were completed during May 1994.
6.2.3 Fuel Cell Installation and Checkout on Natural Gas
The fuel cell and cooling module were delivered and set on their concrete pads at the Penrose site on June
7th, 1994. A photograph showing the fuel cell being lowered onto the pad is shown in Figure 6-1.
-58-
-------
International Fuel Cells
FCR-13524E
PACIFIC ENERGY CORPORATION
POWER PRODUCTION BUILDING
LANDFILL GAS
CLEANUP SYSTEM
WCN15013
FUEL CELL
POWER PLANT
Figure 6-1. Installation ofPC25 at Los Angeles Landfill
Mechanical and electrical installation of the fuel cell and cooling module were completed during July. The
mechanical work included installation of piping to bring clean landfill gas from the GPU to the fuel cell,
installation of natural gas lines to the fuel cell, and installation of coolant piping from the fuel cell to the cool-
ing module. The electrical installation work included connection of the power plant to the grid, installation
and hookup of the electrical metering cabinet and meters, (provided by the L.A. Department of Water and
Power), installation of the telephone wire to the fuel cell controller modem (to enable remote monitoring of
the power plant by ONSI and IFC) and installation of power lines from the fuel cell to the cooling module.
A layer of gravel was added to the site to control dust. A photograph showing the completed installation is
in Figure 6-2.
After completion of the installation, and inspection by the L.A. City Building and Safety Department, ONSI
Corporation initiated startup and checkout of the fuel cell on natural gas. This step was taken to verify proper
operation of the fuel cell prior to modification for landfill gas.
The fuel cell power plant was successfully started on natural gas and operated for a total of 113 hours in two
runs. During this time safety shutdown tests were demonstrated for the City of Los Angeles Building and
Safety Department to demonstrate safe operation in the event of loss of electricity to the power plant and safe
shutdown in the event of a loss of fuel to the power plant, loss of water, or over temperature in the steam
accumulator. Also, during this operating time, stack performance plus system temperatures and pressures
were checked against factory test data and found acceptable, except for a slightly low temperature in the fuel
processor. A thermal control valve was replaced, and the correct temperature was obtained during the second
run.
-59-
-------
International Fuel Cells
FCR-13524E
WCN15074
Figure 6-2. Photograph of the GPU and Power Plant Installed at the Penrose Site
6.2.4 Modifications for Landfill Gas
The fuel cell modifications for operation on landfill gas are described in Section 4.3. The modifications were
installed on site under the direction of an IFC field engineer, from modification kits designed and fabricated
by IFC. The modifications included a larger fuel control valve and fuel flow venturi, a new process fuel re-
cycle orifice, a new cathode exit orifice, and new redundant start fuel shut off valve for the reformer burner.
The modification kit installations were completed in early November.
6.2.5 Checkouts for Landfill Gas Operation
Fuel cell checkouts on landfill gas were initiated in November. Prior to the first startup of the fuel cell on
landfill gas, the GPU was checked out and tested for contaminants. Results of the GPU exit gas samples taken
in October verified that after over 800 total hours of operation, the GPU was still producing gas far cleaner
than originally specified for the fuel cell. Samples were taken during the last hour of the "make" cycle, just
prior id regeneration, (after at least seven hours on line). A sample was taken from the even number beds
(DAB 104 and CAB 106) on October 16th and from the odd number beds (DAB 105 and CAB 107) on October
27th. The results, summarized in Table 6-1 show total sulfur averaging under 0.050 ppmV and total halides
at just 0.003 ppmV. The only contaminant species detected were carbonyl sulfide, methylene chloride, and
trace quantities of xylene, detected at approximately 0.001 ppmV.
The fuel cell was started and operated on clean landfill gas from the GPU for the first time on November 17,
1994. Gross power of up to 80 kW was obtained while checkouts were performed at idle (0 net kW to the
grid) for slightly over two hours. During this initial run, two operating issues were identified with the fuel
control valve: control system instability; and lower than anticipated capacity. Corrections were identified
for both issues, and were installed in the fuel cell in early December.
-60-
-------
International Fuel Cells
FCR-13524E
Table 6-1. GPU Validation Test Results Prior to Start of Fuel Cell Demonstration Field Test
Sample Loca-
tion
GPU Exit ("A" Beds,
DAB104/CAB106)
on line
GPU Exil <"B" Beds,
DABI05/CAB106)
on line
Sample ID
9405858
9405849
9405849 Dup <*>
9405874
9405875
9405875 Dup <4>
Date
10-26-94
10-26-94
10-26-94
10-27-94
10-27-94
10-27-94
Time
12:05
12:14
12:14
N.A
N/A
N/A
Total S
asH2S
0.041 ppmV(')
0.052 ppmVC)
0.051 ppmV
0.046 ppmV ซ!
Total Ha-
lide
asHCL
0.003 ppmV <ป
0.003 ppmV <2>
0.002 ppmV ซ>
0.003 ppmV <2>
0.003 ppmV m
0.002 ppmV <2>
Other
0.001 ppmV Toluene
0.00 1 ppmV Toluene
0.001 ppmV Toluene
0.001 ppmV Toluene
0.001 ppmV Toluene
0.001 ppmV Toluene
NOTES:
( 1 > All as carbonyl sulfide (COS)
(2) All as methylenc chloride (dichloromcihane)
(3) Limit <3 ppmV total sulfur, <3 ppmV total halide
(4) Laborator>' duplicate
(5) All samples taken during last hour of "make" cycle, with approximately 800 total run hours on GPU
6.3 Demonstration Test Results
6.3.1 GPU Performance
6.3.1.1 Operation and Reliability - GPU operating experience during Phase III is summarized in Table 6-2.
During Phase III, the GPU ran 1681 hours in 26 runs, with the longest run being 342 hours. Total GPU opera-
tion including Phase II is 2297 hours. The causes of GPU system shut downs have been identified and cor-
rected, and no outstanding operational issues exist in the GPU at the present time.
Reliability of the GPU has improved as the causes of the shut downs were identified and corrected, so that
extended operation of the GPU should now be possible. Five of the initial six shutdowns were caused by
known or suspected loss of the flare UV flame sensor signal. This was corrected by adjusting the flare time-
out relay. Shut down of runs 7,14,15 and 24 were caused by loss of coolant temperature control in the d-limo-
nene loop. This loss of temperature control is caused by ice build-up in the d-limonene system, due to water
vapor entering the d-limonene system through the air vent in the d-limonene surge tank. Short term, this issue
was corrected by installing a dryer cartridge on the air vent to prevent further water vapor ingestion, and by
instituting an "on the fly" de-icing procedure. The long term solution would include the vent dryer, plus
draining and drying the d-limonene system and recharging with dry d-limonene. Run 8 shutdown was due
to a condenser tank overfill which was attributed to a high (40 gallons in several hours) condensate influx
at the site. No corrections were made and this failure did not reoccur. Run 13 experienced a lockup to the
PLC controller due to a control valve position switch being out-of-limits. The valve limit switch was reposi-
tioned and tightened, and the PLC controller program was modified to prevent the program lockup from reoc-
curring. Run 25 shut down due to loss of Hare flame signal. This was traced to a loose ultraviolet sensor.
This was corrected by reseating the sensor in the socket. The remaining 13 shut downs included six voluntary
shut downs plus seven site related shut downs due to loss of power, or loss of landfill gas pressure to the GPU.
In summary, the GPU System shutdown causes have been identified and corrected, and no known outstanding
issues exist in the GPU at the time of last shutdown.
The GPU adjusted reliability during the demonstration test period, (December 7,1994 through February 19,
1995) was 87.3% (see discussion in Section 6.3.2.1).
-61-
-------
International Fuel Cells
FCR-13524E
Table 6-2. GPU Run Summary
Start Date
Thni March
1994
Oct. 11, 1994
Oci. 12
Oct. 13
Oct. 14
Oct. 17
Oci. 20
Oct. 21
Nov. 3
Nov. 8
Nov. 9
Ncn. 10
Nov. 14
Nov. 28
Dec. 13
Dei 14
Jan 9. 1995
Jan 10
Jan 12
Jan 16
Jan 17
Jan 19
Jan 19
Jan 23
Feb2
Febl6
Run
No.
^
l
1
3
4
5
6
7
8
9
10
II
12
13
14
15
16
1
18
19
20
21
22
23
24
25
26
Run
Hours
9.5
7.7
7.5
5
22.3
18
167
11
3.5
19
16
130
82
342
5
24
20
29
91
21
45
6
20
177
319
83
Total
Hours
616
626
634
641
646
668
686
853
864
878
887
903
1033
1115
1457
1462
1486
1506
1535
1626
1647
1692
1698
1718
1895
2214
2297
Reason for
Shut Down
Flare U.V. sensor
N/A
Step 2 timeout
Flare U.V. sensor
Flare U.V. sensor
Flare U.V. sensor
Flare U.V. sensor
First stage condensate tank overfill
Loss of LFG pressure
Voluntary
Penrose power failure
Program slopped ai 1033 hours but
flare and refrigerator stayed on ai 25
CFMflow
Voluntary
Hi d-limonene surge lank tempera-
lure
High d-limonene surge tank temper-
ature
Voluntary
Voluntary
Penrose breaker (rip
Penrose breaker trip
Voluntary
LFG pressure loss @ Penrose
Penrose breaker trip
Voluntary
Hi d-limonene tank temp.
Loss of flare flame signal
Loss of flare flame signal due to loss
of gas pressure after Bradley S/D
Corrective Action
Taken to GPU
Cleaned dust off sensor
Restarted
Checked Step 2 valve position switches - all ok
Restart with U.V. jumpered
Increase flare timer from 15 to 32 seconds
Increase LFG inlet regulator pressure from 47
psig to 55 psig
No action taken due 10 high LFG condensate
(40 gal. in one day)
No action taken. S/D due to high LFG conden-
sate (40 gal. in one day)
None
None
See below
Tightened loose collar on valve 138 positioner
switch, checked and lightened all other valve
positioner switches
Installed new PLC program
None
None
None
None
Installed new PLC program (PTU7)
None
* None (reset current trip for Penrose breaker
from 5 sec to 10 sec
None
None
* None (checked out current trip and ground fault
trip for Penrose breaker)
Serviced refrigerator (added R-22)
Thaw out refrigerator overnight and restart
Reseated loose U.V. detector
None
-62-
-------
International Fuel Cells
FCR-13524E
6.3.1.2 GPU Contaminant Removal Performance - The GPU consistently removed contaminants in the
landfill gas to levels significantly below the initial goals of < 3 ppmV total sulfur and < 3 ppmV total halides
for over 2200 hours, as shown in Figure 6-3. All Phase III sample were taken by TRC Environmental, during
the last hour before regeneration, per the Test and Quality Assurance Project Plan in Appendix G. The GPU
data is summarized in Appendix D (for Phase II Field Testing), Table 6-1 (Phase III Pre-Start Check), Table
6-3 (Phase III testing), and Appendix H (Phase III testing).
3.0 '
2.5
2.0
PPMV rs "
1.0 '
O.S *
0.4
0.3
0.2
0.1
o.o
MAX ALLOWABLE TOTAL SULFUR <3 PPMV
MAX ALLOWABLE TOTAL HALIDES <3 PPMV
.. Phase III Demonstration Test with
K0y Fuel Cell /Dec. 7. 1994 to Fab. 19. 1995)
Q Total Sulfur (As H2S)
rjj Total Halides (As HCI)
Phase II Fioid Testing (October 20 to 22, 1993)
\ samples)
I Total halides non-detected to 0.032 ppmV (6
I samples)
1 Phase III Pre-Start Check (OcL 26 & 27, 1994)
I Total sulfur 0.046 to .052 ppmV (4 samples)
I
\ Total halides 0.002to0.003ppre(V (4 samples)
i i i i i i i i iii i i i i
Total sulfur non-detected to 0.3B5 ppmV
(7 samples)
Total halide non-detected toO.009 ppmV
(7 samples)
1
1
1
!
Switched to Fresh H2S Removal
Bed February 15, 1995
i i
i(s?i^pj| fsEj |(5J|
0.0 100 200 300 400 100 MO TOO 800 90010001100 1200 1300 1400 1SOO 1600 1700 1800 19002000 21002200 2300
TOTAL GPU OPERATING HOURS 13524-20
952104
Figure 6-3. GPU Exit Contaminant Concentration vs. Time
The GPU has consistently removed total halides (as HCI) from inlet levels of 45 to 60 ppmV in the raw landfill
gas to very low or undetectable levels at the outlet. During the Phase II Field Testing on October 20 to 22,
1993, with 200 hours on the GPU, four samples showed no detectable halides (0.002 ppmV detection limit),
while one sample tested at 0.008 ppmV and one sample tested at 0.032 ppmV. During the Phase III Pre-Start
Testing, four samples showed detectable total halides averaging 0.048 ppmV. During the Phase III testing
the sample at 1685 hours tested at 0.009 ppmV total halides. All six (6) remaining samples out to 2235 hours
showed no detectable halides (individual species detection limits 0.001 to 0.020 ppmV). The excellent halide
removal performance of the GPU allowed IFC to eliminate the addition of a halide guard bed in the PC25
power plant, as was originally planned.
Total sulfur (as F^S) was reduced from about 110 ppmV (about 10 ppmV from organic sulfur, plus 100 ppmV
H2S) to between non-detectable to 0.385 ppmV. The only sulfur species detected was carbonyl sulfide. The
elevated levels of 0.173 to 0.385 ppmV of carbonyl sulfide measured on February 9th and 10th, 1995 are
believed due to a slight increase in H2S exiting the non-regenerable H2S removal bed, since hydrogen sulfide
at the H2S bed exit was measured at 1.0 to 2.7 ppmV on February 14th. Earlier laboratory work at IFC showed
that H2$ is converted to carbonyl sulfide over the activated alumina in the downstream drier bed by the reac-
tion: H2S + CC>2 = COS + H2O, due to the removal of the product water by the alumina. The resulting carbo-
nyl sulfide is not readily removed by the low temperature carbon bed. The non-regenerable H2$ removal bed
-63-
-------
International Fuel Cells
FCR-13524E
was switched over to a fresh bed on February 15, 1995, and the exit H2$ level returned to non-detectable.
The carbonyl sulfide level measured shortly after, on February 17th, also fell to just 0.061 ppmV.
Based on the Penrose operating experience, the useable life of the hydrogen sulfide removal beds is 21 days,
which yields an apparent capacity of 12 grams of sulfur per gram of carbon in the bed. The small, 119 liter
bed volume of the two beds used in the demonstration test was selected for proof of principle only. The com-
mercial installation would use large, commercial tanks designed for low cost, ease of servicing, and long
changeout times.
Table 6-3. GPU Contaminant Removal Performance During Phase III
SAMPLING DATE (1995)
Total GPU Operating Time (Hours)
Sampling Time
GPU Process Counter
SULFUR COMPOUNDS (ppmV)
hydrogen sulfide
methyl me reap tan
ethyl rnercaptan
dimethyl sulfide
dimethyl disulfide
Carbonyl sulfide
carbon disulfide
Total Sulfur
VOLATILE ORGANIC COMPOUNDS (ppmV)
dichlorodifluorome lhahe
1 . l-dichloroethane
benzene
chlorobenzene
ethyl benzene
methylcne chloride
styrene
trichloroethene
toluene
tetrachloroethene
vinyl chloride
xylene isomers
cis- 1 .2-dichloroethene
Total Halides as Cl
Jan 19
1685
17:00
24969
Jan 20
1701
09:22
24900
<0.004
<0.004
<0.004
<0.004
<0.002
<0.004
<0.002
nd
<0.02
<0.001
0.001
<0.00l
<0.00l
0.005
<0.001
<0.001
<0.002
<0.001
<0.002
0.001
<0.001
0.009
<0.004
<0.004
<0.004
<0.004
<0.002
<0.002
<0.002
nd
<0.02
<0.001
<0.002
<0.001
<0.00l
<0.002
<0.001
<0.001
0.003
<0.001
<0.002
0.003
<0.001
nd
Jan 25
1710
16:14
53080
<0.004
<0.004
<0.004
<0.004
<0.002
0.071
<0.002
0.071
<0.001
<0.001
<0.002
<0.001
-------
Internationa] Fuel Cells
FCR-13524E
6.3.1.3 GPU Exit Gas Heat Content - The heat content of the GPU exit gas was measured lo provide an
accurate basis for fuel cell efficiency calculations. The average GPU exit gas heat content was determined by
averaging the hourly on-line gas chromatograph data provided by Pacific Energy for the raw landfill gas at the
GPU inlet, and adjusting this using a correction factor. The correction factor was based on a comparison of six
GPU exit gas ASTM heat content measurements which were compared to the Pacific Energy hourly average
reading for the inlet gas taken at the same time. The GPU exit gas heat contents averaged one percent higher
than the inlet samples taken at the same time, so a correction factor of 1.01 was applied to the on-line inlet gas
data. The summary of the correction factor data and calculation is given in Table 6-4. Additional heat content
data is given in Appendix H (see Section 3.5 and Appendix B).
Table 6-4. GPU Exit Gas Heat Content
Comparison of ASTM Method Heat Content Measurements on Treated GPU Exit Gas
to On-Line Raw Landfill Gas Heat Content Measurements
Penrose Landfill - Phase III Fuel Cell Energy Recovery Demonstration
January 19 - February 17, 1995
SAMPLING DATE
SAMPLING TIME
Treated Landfill Gas Composition Measured
By ASTM Method at GPU Exit (%)
nitrogen
carbon dioxide
methane
ethane
propane
iso-buianc
iso-peniane
n-pentane
hcxancs
heptanes
GPU Exit HHV by ASTM Method
Biu standard cubic fool
Kcal'siandard liter
GPU Exit LHV by ASTM Method
Bin standard cubic fool
Kcal standard liter
GPU Inlet HHV by Pacific Energy
On-Line Analyzer
HHV (Btu standard cubic foot)
HHV (KcaL'standard liter)
Heat Content Correction Factor
[GPU Exit HHV/GPU Inlet HHV]
Jan 19
16:44
Jan 20
09:27
Jan 25
16:09
Jan 26
08:31
Feb9
10:37
FeblO
09:26
Febl7
13:33
16.266
35.542
44.165
0.024
nd
nd
nd
nd
nd
nd
17.251
38.896
43.807
0.029
nd
nd
nd
nd
nd
nd
16.244
39.555
44.142
0.049
nd
nd
nd
nd
nd .
nd
r 16.34
39.531
44.092
0.037
nd
nd
nd
nd
nd
nd
23.888
36.042
40.07
nd
nd
nd
nd
nd
nd
nd
17.656
38.863
43.481
nd
nd
nd
nd
nd
nd
nd
20.096
34.908
44.996
nd
nd
nd
nd
nd
nd
nd
446
3.97
402
3.58
443
3.94
447
3.98
446
3.97
405
3.60
439
3.91
454
4.04
399
3.55
402
3.58
401
3.S7
364
3.24
395
3.52
409
3.64
437
3.89
435
3.87
445
3.96
445
3.96
436
3.88
429
3.82
no da la
no data
1.02
1.02
1.00
1.00
0.93
1.02
no data
NOTES:
1 . nd - non-detected
2. Standard Conditions at 20 ฐC
3. Average correction factor is 1.01 (Exclude Feb. 9 data from average-suspected sampling error.)
-65-
-------
International Fuel Cells
FCR-13524E
6.3.2 Fuel Cell Performance
6.3.2.1 Fuel Cell Operation and Availability - The fuel cell operation on landfill gas is summarized in
Table 6-5. Checkouts on landfill gas were conducted in November 1994. During this time the fuel cell was
started twice for a total of two hours, and issues with the fuel control valve capacity and stability were identi-
fied and corrected.
The field test began with Run No. 3 on December 7,1994 until the end of Run No. 10 on February 19,1995.
During this period the fuel cell operated for 707 additional hours on landfill gas with no forced outages. Only
one of the eight shutdowns was due to the fuel cell. Of the eight shutdowns, four were due to site related
causes (one power loss due to Penrose breaker trip, one loss of landfill gas pressure when the Penrose power
station was shut down for maintenance, and two by bad gas when the Bradley landfill went off-line). Three
shutdowns were due to the GPU (two due to refrigeration over temperature and one due to a loose flame sensor
on the flare), and one shutdown (Run No. 9) was due to a bad sense module in the fuel cell control system.
liable 6-5. Summary of Fuel Cell Operations on Landfill Gas
Type
Opera-
lion
check-
outs
4
Tc
eld
SI
*
Run*
Start
Time
Date
Shut-
down
Time
Date
Run
Hours
Total
LFG
Hours
Reason
For
Shutdown
Corrective
Action
Taken to
Fuel Cell
1
2
11:19
14:56
1 1/16/94
11/17/94
11:19
17:05
I
11/16/94
1 1/17/94
0
2
0
2
Frozen fuel control
valve caused valve
motor fuse to blow
Unstable fuel flow
control, and insuffi-
cient fuel flow
Freed stuck fuel
control valve
* Replaced fuse
Replaced 1/2" fuel
control valve with
1 " valve & adjusted
controller
3
4
5
6
7
8
9
10
12:16
10:28
15:12
10:32
11:55
12:45
08:00
18:50
12/7/94
1/14/95
1/17/95
1/23/95
2/3/95
2/6/95
2/16/95
2/16/95
16:02
08:23
07:19
17:01
15:35
19:12
08:15
08:38
12/12/94
1/16/95
1/19/95
1/30/95
2/4/95
2/15/95
2/16/95
2/19/95
124
46
40
175
27
223
0
72
126
172
212
387
414
637
637
709
GPU shutdown
Penrose breaker
trip
Penrose shutdown
for maintenance
GPU shutdown
Bad gas from Brad-
ley Landfill
GPU shutdown
Intermittent failure
of inverter cooling
fan sense module
Bradley landfill
shut down
None required
Replaced leaking
feed water shut off
valve during shut-
down
None required
None required
* None required
None required
None required
Replaced bad fan
sense module
None required
-66-
-------
International Fuel Cells FCR-13524E
The fuel cell was operated at up to 137 kW, which is 3 kW below the goal for operation on landfill gas. Mea-
surements taken on the landfill gas fuel cell indicate that the standard natural gas fuel pump used in the power
plant (a steam ejector) is not providing the anticipated suction. This deficiency can be overcome by modify-
ing the ejector specifically to provide the required suction using landfill gas. The recurring cost for a modified
ejector would be essentially the same as the standard natural gas ejector.
An endurance operating condition of 120 kW was selected for the bulk of the field test operation to provide
a margin for steady fuel cell operation during periods of sub-standard gas quality, which occur periodically
due to upsets in gas quality from the active landfill (Bradley) which supplies gas to the Penrose site.
The fuel cell adjusted availability was completed per the Quality Assurance Project Plan (Appendix G pg
16).
Fuel cell availability is adjusted to compensate for factors which are not caused by the power plant,
as follows:
Raw availability (OPERATING TIME divided by elapsed clock time since first start) is adjusted
to account for
unforced outages not due to power plant
- shutdowns due to operator error
- waiting time for replacement parts where parts were recommended the customer have on
hand
periods of time when power plant could be worked but manpower not available (week-
ends, vacations)
OPERATING HOURS
Adjusted availability = -; ;: ,
[(elapsed clock time) - adjustment]
During the test period, the only unavailable time due to the fuel cell is the 10.6 hours between the shutdown
of run No. 9 due to the failed sense module at 08:15 on February 16th, to the startup of Run No. 10 at 18:50.
This yields an adjusted availability 707 = gg ,.
1782.3-1064.7
The GPU adjusted availability was also computed for the same test period. The availability of the fuel cell
and GPU was segregated to provide direct comparison of the landfill gas powered fuel cell to the measured
reliability of the commercial natural gas fueled fleet of fuel cell power plants. The separate availabilities also
recognize the fact that the fuel cell is a commercial piece of equipment, while the GPU is a first-of-a-kind
experimental demonstration unit whose main objective is to demonstrate the critical technical and operational
issues for the commercial landfill gas cleanup system.
During the test period the GPU experienced three shutdowns that resulted in lost time. The first shutdown
on December 13 at 1602 was caused by high temperature in the d-limonene tank, and resulted in 46 lost hours
until the GPU was checked out and ready to provide gas to the fuel cell. The second shutdown at 1700 on
January 23 was also due to high d-limonene tank temperature, and caused 43.7 hours of lost time. The third
shutdown on February 15 at 1915 was caused by a loose UV detector and resulted in 13 hours of lost time.
The total lost time during the test period was 102.7 hours.
The adjusted availability for the GPU during the test period is therefore:
707 (1782.3 - 972.6) = 87.3%
6.3.2.2 Fuel Cell Power Plant Efficiency - The fuel cell efficiency was calculated over two periods during
the field test. The first period covered six days from January 24,1995 through January 30,1995. Efficiency
during this six day period of continuous operation was 37.1 percent. The second period covered eight days
-67-
-------
Internationa] Fuel Cells
FCR-13524E
from February 9th, 1995 through February 17,1995. Average efficiency for this eight day period, which in-
cluded a brief shutdown, was 36.5 percent. The details of the calculation are given in Table 6-6.
Table 6-6. Fuel Cell Electrical Efficiency on Landfill Gas
Penrose Landfill - Phase HI Fuel Cell Energy Recovery Demonstration
' January 24 - February 17, 1995
Period
1/24/95 (0707) 10 1/30/95 (1023)
2/9/95 (1 102) to 2/17/95 (0733)
Energy
Output
(LADWP Meter)
(kWh)
16800
18400
Gas
Consumption
(Yokagawa Meter)
(SL)
I.I1E + 07
1.26E + 07
Lower
Healing
Value
(Kcal/SL)
3.SO
3,45
Energy
Input
(Real)
(Real)
3.894E + 07
4.33E + 07
Efficiency
37.1%
36.5%
NOTES:
1 . Heating value data is from Pacific Energy's on-line raw gas analyzer HHV hourly averages corrected to GPU exit LHV. A correction
factor ( 1 .01) was developed from a comparison of six GPU Exit ASTM measurements to six GPU Inlet HHV on-line averages. The HHV
was then convened to the LHV using the correction factor 0.900. The following equation was used for the complete conversion:
Exii LHV - GPU Inlet HHV x 1,01 x 0.900
2. Efficiency - Energy Output (kWh) x 860.5 Kcal/kWh x 100
Gas Consumed (SL) x LHV (Kcal/L)
SL - standard liters at 15.5ฐC
The fuel cell power plant electrical efficiency was measured by recording output kW hours at Location ฉ
on Figure 6-4 and dividing by the input landfill gas lower heating value at Location Qi) on Figure 6-4. The
electric power was measured using a utility grade meter calibrated by LADWP. The landfill gas to the fuel
cell was measured using a Yokogawa YFCT Flow Computing Totalizer (Style B). The determination of the
heating value of the landfill gas is discussed in Section 6.3.1.3. Additional details of the data and calibrations
is given in Section 3.1 of Appendix H.
Gas
Pre-treatment
Unit
(GPU)
Fuel Cell
Power Plant
Output
Performance Demonstration Interfaces:
Emission Test Interfaces:^), ฎ, (
FCR13S24C
Figure 6-4. Demonstrator System Schematic
13524-21
952104
-68-
-------
International Fuel Cells
FCR-13524E
6.3.2.3 Fuel Cell Maintenance and Operator Requirements - The operation and maintenance cost factors
for the landfill gas fuel cell demonstration are compared with the PC25A commercial fuel cell experience in
Table 6-7. Based upon a comparison of the operation and maintenance cost factors after 707 hours, the O&M
costs for the landfill gas fuel cell power plant project to be comparable to slightly higher than for natural gas,
depending upon Nitrogen content of the fuel. A longer period of steady endurance operation on landfill gas is
desirable to develop a firmer basis for long term O&M costs for the LFG fuel cell.
Operation factors for the landfill gas fuel cell demonstration including startup, operation and availability are
comparable with commercial natural gas experience. The efficiency based on fuel lower heating value is 37
percent for the landfill gas at 120 kW, This efficiency is in line with the Phase I projection of 36.4 percent
at 200 kW shown in Table 3.5. The lower efficiency requires about 8 percent higher fuel flow, for the landfill
gas fuel cell, but this is offset by the lower fuel costs for landfill gas. The demonstrated maximum output
of the PC25 A power plant is 137 kW, with a steady state output of 120 kW. This output must be increased
to 200 kW steady output using landfill gas to achieve comparable operating costs. The means to achieve a
200 kW rating are discussed in more detail in Section 7.
The maintenance cost factors for unscheduled maintenance project to comparable cost using landfill gas fuel,
based upon the limited mean time between forced outages and availability data generated in the 707 hours
of operation on landfill gas. No scheduled maintenance has been performed to date, but the landfill gas power
plant scheduled maintenance is anticipated to be the same as the natural gas power plant.
Table 6-7. Operation and Maintenance Cost Factors for Commercial Applications
Factor
PC25 A Natural Gas
Fuel Cell
Commercial Experience
PC25 A Landfill Gas
Fuel Cell
Demonstration
Comments
Operation
Sunup from energized off
Normal operation
Availability
Rated ouipui
Efficiency (LHV)
Heat recovers'
Fuel Hcaimg Value
HHV
LHV
5 hours or less
Unattended, automatic
95%
200 kW
40%
1 92.000 kcal/hr@ 200 kW
Natural Gas
8.72-1 0.68 Kcal/SL
7.86-9.62
5 hours or less
Unattended, automatic
98.5%
!20kW(l37kWmax)
37ซr
Not demonstrated
LFG
3.92 kcal'SL
3.53 kcal'SL
LFG utilizes electric
start option
PC25 C can be modified
to make 200 kW on LFG
Lower efficiency re-
quires 8% higher (uel
flow, but this is offset by
lower fuel costs lor
Landfill Gas
Projected heat recovery
208.000 kcal/hr on LFG
Penrose LFG heating
value at low end of the
range
Maintenance
Scheduled
Unscheduled
MTBFO '
availability
@ 2.000 hours
(during operation)
@ 8.000 hours
(while shutdown)
2.600 hours
95%
Not demonstrated
Not demonstrated
None in 707 hours
98.5%
Projected same as nat-
ural gas
Operation to date indi-
cates LFG fuel cell com-
parable to natural gas
experience
Notes: ' Mean lime between forced outages
-69-
-------
International Fuel Cells
FCR-13524E
6.3.3 Emissions
The fuel cell power plant emissions at 120 kW on landfill gas are summarized in Table 6-8. The average
emissions are as follows: NOX = 0.12 ppmV; Sulfur Dioxide = non detectable (0.23 ppmV detection limit);
and Carbon Monoxide = 0.77 ppmV. All readings are reported as parts per million, dry gas, corrected to 15
percent Oi- The results are also presented as a mass emissions rate in grams per hour, and grams per kilowatt
hour.
Table 6-8. Fuel Cell Emissions Summary on Landfill Gas
Penrose Landfill Phase III Fuel Cell Energy Recovery Demonstration
February 17, 1995
SAMPLING TIME Measurement
Melhod
EMISSION CONCENTRATION
(actual dry measurements)
nitrogen oxides (ppmV) A
sulfur dioxide (ppmV) B
carbon monoxide (ppmV) C
oxygen (%) D
carbon dioxide (%) E
EMISSION CONCENTRATION
(dry measurements corrected to 15% oxygen)
nitrogen oxides (ppmV)
sulfur dioxide (ppmV)
carbon monoxide (ppmV)
VOLUMETRIC FLOW RATE (dscm/m)
STACK TEMPERATURE (eC)
MASS EMISSION RATE (grams/hour)
nitrogen oxides
sulfur dioxide
carbon monoxide
MASS EMISSION RATE
(grams/kilowatl-hr)
nitrogen oxides
sulfur dioxide
carbon monoxide
0800-
0900
0.3
<0.5
1.5
7.96
12.5
0.14
<0.2
3
0.68
10.1
56.7
0.35
<0.8
0
1.06
0.00
29
<0.0
067
0.00
88
0950-
1050
0.17
<0.5
1.8
8,01
12.6
0.08
<0.2
3
0.82
10.1
56.7
0.20
<0.8
0
1.27
0.00
16
<0.0
067
0.01
06
1155-
1255
0.31
<0.5
2.1
7,88
12.7
0.14
<0.2
3
0.95
9.4
43.3
0.33
<0.7
5
1.37
0.00
28
<0.0
062
0.01
15
1332-
1442
0.17
<0.5
2.3
7.8
12.3
0.08
<0.2
3
1.04
9.4
43.3
0.18
<0.7
5
1.51
0.00
15
<0.0
062
0.01
25
1457-
1SS7
0.41
<0.5
0.6
8.03
12.4
0.19
<0.2
3
0.28
9.7
42.8
0.46
<0.7
8
0.41
0.00
38
<0.0
065
0.00
34
1622-
1722
0.18
<0.5
1.9
7.91
12.5
0.08
<0.2
3
0.86
9.7
42.8
0.20
<0.7
8
1.29
0.00
17
<0.0
065
0.01
07
Average
0.26
-------
International Fuel Cells FCR-13524E
The fuel cell emissions data are based upon six, one-hour continuous monitor measurements conducted on
February 17,1995. The continuous emission monitors were calibrated before and after each test for zero and
span drift. The details of the test procedures and data analysis are discussed in more detail in Appendix H.
The results of the emissions testing conducted over the six, one-hour periods are believed to be representative
of longer term continuous emissions, since the fuel cell power plant controller continuously adjusts the fuel
and air to maintain constant temperature inside the reformer burner where the carbon monoxide and NOX are
generated. In addition, the results for the landfill gas power plant are in good agreement with other emissions
data measured on PC25 A power plants. The average results for 16 fuel cell power plants tested at the factory
using natural gas fuel are: 0.46 ppmV NOX, and 1.1 ppmV CO. Emissions tests were also conducted on a
PC25 A sited at the SCAQMD Headquarters building in Diamond Bar, CA. The results showed 0.45 ppmV
NOX and 1.1 ppmV carbon monoxide. The Diamond Bar site results were confirmed by two independent
laboratories, and were used by the agency as the basis for a blanket exemption from air permit requirements
for fuel cells in the Los Angeles basin. The slightly lower NOX emissions for the landfill gas power plant
are likely a result of the lower reformer burner flame temperatures associated with the low heating value of
the landfill gas fuel.
6.3.4 Quality Assurance
A summary of the quality assurance goals and test results is given in Table 6.9. The goals for accuracy and
precision are based on the QAPP in Appendix G, and the quality assurance test results are based on the test
data summarized in Appendix H. Typical concentrations, detection limits, and blank results for targeted com-
pounds in the raw landfill gas at the Penrose Landfill are summarized in Table 6.10.
The quality assurance measurements for accuracy of hydrogen sulfide and for 3 of the 4 tested haJogenated
volatile organic compounds did not meet the 15% goal. The GPU removed all of these selected compounds
to below the detection limit in the GPU exit gas, so these errors are not significant to the conclusions regarding
the overall effectiveness of the GPU for sulfur and halide removal. The 30.7% high reading for hydrogen
sulfide could lead to an overstatement of the apparent sulfur capacity of the hydrogen sulfide removal bed
except that a nominal 100 ppmV value (based on historical Penrose data, in Table 4-4) was used for these
calculations. The 100 ppmV H2S value was corroborated by Drager tube measurements taken at the site dur-
ing the demonstration test.
The accuracy of the ASTM D3588-91 method did not meet the 2% goal for all constituents, most notably
methane at -3.5%. All other hydrocarbon species were negligible in these tests. The impact of the 3.5% error
could be an overstatement of the apparent fuel cell efficiency by 3.5% (e.g., the reported 37.1% efficiency
could be 35.8%). The close agreement between heating value measured by the Pacific Energy on-line analyz-
er and the ASTM method indicates the real error is probably less than 3.5% (see Table 6-4 on page 65).
The quality assurance tests of the emissions monitors showed that the SO2, CO, CO2 and 02 measurements
generally met or bettered the QA goals. The NOX emissions tests exceeded the stated QA goals for accuracy
and precision. Accuracy measured -22.4% and -20.7% vs. the 15% goal. Precision, measured as zero drift
(-28 to 35.2%), and span drift (-32 to +21.5%) failed to meet the 10% goal. The NOX QA results indicate
a higher degree of uncertainty in the NOX data, which is not surprising due to the low average value of just
0.12 ppm. An attempt was made to minimize the impact of zero and span drift by recalibrating the thermo-
electron Model 10A chemiluminescent NO/NOX analyzer before each 1 hour continuous analysis period, and
using the average for the six periods. Even a 100% error in the final result would not change the conclusion
that the exhaust NOX level of the fuel cell operating on landfill gas is exceptionally low.
-71-
-------
International Fuel Cells
FCR-13524E
r Table 6.9 Summary of Quality Assurance GoaJs and Tfest Results
i
2
3
4
Measurement
Sulfur Compounds
Hydrogen Sulfide
Volatile Organic
Compounds
Vinyl Chloride
cis-l.
2-dichloroethene
l.l-dichloroe thane
teirachloroeihenc
GPU Input Gas
Heat Content
GPU Output Gas
Heal Content
Method
EPA16&18
EPA -TOM
On-line
Analyzer
ASTM
D3588-9I
Operating
Range
(A)
0-102ppmV
(A)
O.I-l.4ppmV
3.9 - 5.9 ppmV
1.2-2.9ppmV
2.4 - 4.8 ppmV
N/A
3.56- 4.09 kcal/sl
PRECISION
Goal
5%
15%
15%
15%
15%
2%
2%
Fuel Cell Exhaust Emissions
5
6
7
B
9
SO? Emissions
NO, Emissions
CO Emissions
CO; Emissions
O? Emissions
EPA-6C
EPA-7E
EPA- 10
EPA-3A
EPA-3A
0-100ppmV
0-2JppmV
0-100ppmV
0-25%
0-25%
5%
10%
10%
5%
5%
Results
0.6%
19.0%
5.8%
6.9%
6.4%
(B)
(B)
(B)
(B)
(B)
N/A
0.11%
Zero
Drift
-2.1 to
+.9%
-28 to
+35.2%
2.810
+ 1. 9%
(AiTypical value in landfill gas ai Penrose- See Table 6.10
(D)
Span
Drift
-1.2 to
+1.3%
-32 to
+21.5%
-30io +
2.1%
ACCURACY
Goal
15%
15%
15%
15%
15%
2%
2%
5%
15%
10%
510%
-4.0% (E)
-22.4%. (E)
-20.7%
-5.4% (E)
1.3% (E)
0.8% (E)
Effect on Data
Conclusions
H2S not detected at GPU
exit, so accuracy not sig-
nificant to conclusion re-
garding GPU effective-
ness.
These species were never
detected at GPU exit, so
effect of not meeting
precision (vinyl chloride)
or accuracy goal (vinyl
chloride, cis-l ,
2-dichloroe thane.
tetrachloroethene) is not
significant to conclusion
regarding GPU
effectiveness
Meets QA goal for accura-
cy
Accuracy does not meet
QA goals for some
species. Net effect on
heal content is possibly
3 10 4%.
Meets QA goals
Low absolute NO, values
make higher uncertainty
less significant
All but 1 span drift meets
QA goals
Meets QA goal
Meets QA goal
(B i Appendix H. Table 1 1 -2. page H-5 1
1C) Relative standard deviation compared with 4 ASTM samples taken within 1 hour. Appendix H. Table ll-l, page H-50
(DtAppendix K. Table ll-l. page H-50
(E ) Appendix H, Table 1 1 -3. page H-53
-72-
-------
International Fuel Cells
FCR-13524E
Table 6.10 Typical Concentrations, Detection Limits, and Blank Results for
Targeted Compounds in the Raw Landfill Gas at the Penrose Landfill
Sulfur Compounds (ppmV)
1. H2S
2. Methyl mercaptan
3. Ethyl mercaptan
4. Dimethyl sulfide
5. Dimethyl disulfide
6. Carbonyl sulfide
7. Carbon disulfide
8. Total sulfur as H2S (ppmV)
Typical Value in
Untreated Landfill Gas
102.0
3.0
0.5
6.5
<0.07
0.2
<0.07
109.0
Detection Limit
Objective
0.04
0.04
0.04
0.04
0.02
0.04
0.02
0.28
Blank Samples
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
<0.002
Volatile Organic Compounds (ppmV)
1 . Dichlorodifluoromethane
2. 1.1 -dichloroethane
3. Benzene
4. Chlorobenzene
5. Ethylbenzene
6. Methylene chloride
7. Styrene
8. Trichloroethene
9. Trichlorofluoromethane
10. Toluene
1 1 .Tetrachloroethene
12. Vinyl chloride
l3.Xylene isomers
14.cis-1.2-dichloroethene
15. Total hal ides as Cl
0.3-0.9
1.2-2.9
1.1-1.7
0.6-1.4
4.5-12.0
4.0-11.0
0.5-1.1
1.3-2.4
0-0.6
28.0 - 47.0
2.4 - 4.8
0.1-1.4
5.0-28.0
3.9-5.9
47.0 - 67.0
0.009
0.002
0.002
0.002
0.002
0.003
0.003
0.001
0.004
0.002
0.002
0.005
0.005
0.003
0.086
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
-73-
-------
International Fuel Cells
FCR-13524E
7.0 PHASE IV GUIDELINES AND RECOMMENDATIONS
The purpose of this phase is to prepare guidelines and recommendations as to how the PC25 C power plant
can be modified to achieve full rated power when operated on landfill gas, based upon the experience gained
testing the PC25 A model in this program.
The PC25 C power plant is designed to produce 200 kW of net power when operated on natural gas having
a higher heating value range of 8.72 kcal/SL to 10.68 kcal/SL. Landfill Gas (LFG) contains significant
amounts of N2 and CO2 which lower the higher heating value. A Landfill Gas with 50 percent methane and
a higher heating value of 4.45 kcal/SL was used for this study. A PC25 C power plant operated on LFG having
4.45 kcal/SL higher heating value would produce a projected 140 kW of net power. To increase the net power
higher flows of LFG would be required to obtain an equivalent natural gas fuel content and heating value.
This phase investigated approaches to achieve increased fuel flows.
Fuel flow in the PC25 C power plant is achieved by using a steam driven ejector. This approach has the benefit
of reducing parasitic power and provides an inherent fail safe feature in that a loss of steam flow will automati-
cally terminate fuel flow and prevent damage to the fuel processor. The injector also has a finite pumping
capacity which requires all components in the fuel delivery system to have small pressure drops.
Increasing fuel flows result in a corresponding increase in pressure drops, so the impact of replacing or modi-
fying components in the fuel delivery system to minimize pressure drop increases was determined. The PC25
C fuel delivery train (Figure 7-1) to the hydro-desulfurizer contains two fuel isolation valves and a check
valve having inside diameters (I.D.) of 2.54cm and a fuel control valve having an I.D. of 1.27 cm connected
with 2.54 cm diameter piping. Increasing the I.D. of the isolation valves and the check valve to 3.81 cm, the
I.D. of the fuel control valve to 2.54 cm, and the connecting plumbing to 3.81 cm (Figure 7-2) would accom-
modate the increased flow without producing unacceptable pressure drops. These changes would not require
major modifications to the power plant.
2.54 cm I.D. 2.54 cm I.D. 2.54 cm I.D. 1.27 cm I.D.
< 2.54 cm Dia. Plumbing ป
13524-22
972 BM
Figure 7-1. PC25 C Fuel Delivery Train
3.81 cm I.D. 3.81 cm I.D. 3.81 cm I.D. 1.27 cm I.D.
3.81 cm Dia. Plumbing ป
13524-23
S72804
Figure 7-2. PC25 C Fuel Delivery Train Modified for Operation on Landfill Gas
-74-
-------
International Fuel Cells FCR-13524E
The high flow rates of landfill gas would also increase the pressure drop across the fuel flow fields within
the cell stack assembly (CSA) and would result in a corresponding increase in the fuel inlet operating pres-
sure. This increased fuel side operating pressure affects the desired pressure differentials between the fuel
and air sides of a cell. This differential is known as "reactant cross pressure". The increase in fuel inlet pres-
sure by itself does not result in an unacceptable increase in fuel delivery system pressure drops, but the corre-
sponding increase in reactant cross pressure exceeds established operating limits. Two approaches were con-
sidered for lowering the fuel inlet pressure: increasing the cross-sectional area of the flow fields and adjusting
the flow directors in the CSA fuel manifolds with a corresponding change in the size of the cathode exit flow
orifice.
The fuel flow-field depths would have to increase by forty percent to obtain an inlet pressure which results
in an acceptable reactant cross pressure. While theoretically possible, this approach could create a manufac-
turing disruption and results in a taller CSA. Both consequences could have a negative cost impact on the
power plant. Slight adjustments to the flow directors in the fuel manifold and a corresponding reduction in
the size of the cathode exit flow orifice would result in a reactant cross pressure which would be acceptable
at flows sufficient to produce 175 kW of power. For flows to produce 200 kW of power the magnitude of
the cathode exit flow orifice size reduction results in the need for a larger cathode air blower. This approach
has a minimum impact on CSA production costs and CSA height, and is recommended as the preferred
choice.
The operating characteristics of the PC25 C ejector at LFG flows required to produce 200 kW was not avail-
able and bench tests were conducted to generate the data. A modified ejector having ten percent larger second-
ary mixing tube was also characterized. The data indicate that the PC25 C ejector is capable of providing
LFG flows sufficient to produce 175 kW of net power but does not have the capacity to provide flows required
to produce 200 kW of net power. The data obtained for the modified ejector indicate that it would be capable
of providing LFG flows required to produce 200 kW of power. The use of a modified ejector would have
a minimal cost impact.
-75-
-------
International Fuel Cells FCR-13524E
8.0 CONCLUSIONS
1. Based on the environmental and economic evaluation of the commercial fuel cell energy system,
there is a large potential market for fuel cells:
The fuel cell landfill gas to energy conversion system provides net reduction in total emis-
sions while simultaneously mitigating the methane from the landfill gas.
With the initial product prices, fuel cells will be competitive in landfill sites located in high
electric cost areas in sites with average commercial rates; where heat can be utilized or where
there is a credit for the environmental reductions from the fuel cell energy conversion sys-
tem.
When the projected mature product price is achieved, fuel cells will be competitive for most
application scenarios. In many situations, fuel cells will provide net revenues to the owners
of the operating landfills. This could, in the long term, result in methane mitigation without
additional cost of any sort to the ultimate consumer.
2. The gas pretreatment unit (GPU) for cleaning landfill gas to fuel cell was successfully designed,
installed, permitted, tested, and validated:
" A permit was granted by South Coast Air Quality Management District for operation in Los
Angeles basin.
Total 2297 hours of operation, including 709 hours operation with the fuel cell.
Adjusted availability of the GPU during the fuel cell demonstration test period was 87.3%.
Documented total sulfur removal betters fuel cell requirements (< 3 ppmV total sulfur).
Documented total halide removal betters requirements for fuel cell (< 3 ppmV total halides).
GPU flare safely disposed of landfill gas contaminants by achieving destruction efficiencies
above 99 percent.
The observed hydrogen sulfide removal bed capacity was 12 grams of sulfur per gram of
impregnated carbon.
3. Fuel cell modifications for operation on landfill gas were successfully demonstrated with a com-
mercially available PC25 A fuel cell power plant:
Operation up to 137 kW.
Efficiency of 37. ] percent at 120 kW.
Exceptionally low secondary emissions (dry gas, corrected to 15 percent 02)
carbon monoxide = 0.77 ppmV
nitrogen oxides = 0.12ppmV
sulfur dioxide = not detected
No forced outages, and adjusted availability of 98.5 percent.
Total 709 hours operation on landfill gas.
-76-
-------
International Fuel Cells FCR-13524E
4. The fuel cell can be modified to operate at rated 200 kW power on landfill gas.
The model PC25 C power plant should be able to generate 200 kW with modifications to
the fuel control plumbing, stack manifolds, ejector, and process air blower, assuming the
land fill gas contains 50% methane (a higher heating value of 4.45 kcal/sl), and an average
sustainable net flow of 2830 scmd (100,000 scfd) of clean landfill gas to the fuel cell.
The recurring cost changes for these modifications is minimal.
5. Additional testing is recommended to demonstrate endurance operation on landfill gas and
provide data for reducing the cost of the gas cleanup system.
-77-
-------
TECHNICAL REPORT DATA
(Please read Instructions on the revene before completing)
1. REPORT NO.
EPA-600/R-98-002a
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Demonstration of Fuel Cells to Recover Energy from
Landfill Gas; Phase III. Demonstration Tests, and
Phase IV. Guidelines and Recommendations*
REPORT DATE
January 1998
. PERFORMING ORGANIZATION CODE
7. AUTHORtS)
J. C. Trocciola and J. L. Preston
8. PERFORMING ORGANIZATION REPORT NO.
FCR-13524E
9. PERFORMING ORGANIZATION NAME AND ADDRESS
International Fuel Cells Corporation
195 Governors Highway
South Windsor. Connecticut 06074
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-Dl-0008
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Air Pollution Prevention and Control Division
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; 1/93 - 4/95
14. SPONSORING AGENCY CODE
EPA/600/13
18.SUPPLEMENTARY NOTES APPCD project officer is Ronald J. Spiegel, Mail Drop 63, 919 /
541-7542. (*) Volume 1. Technical Report. Volume 2 consists of Appendices A-H.
16. ABSTRACT
The report summarizes the results of a four-phase program to demonstrate
that fuel cell energy recovery using a commercial phosphoric acid fuel cell is both
environmentally sound and commercially feasible. Phase I, a conceptual design and
evaluation study, addressed the technical and economic issues associated with oper-
ating the fuel cell energy recovery system of landfill gas. Phase II included the de-
sign, construction, and testing of a landfill gas pretreatment unit (GPU) to remove
critical fuel poisons such as sulfur and halides from the landfill gas, and the design
of fuel cell modifications to permit operating on low heating value (LHV) landfill gas.
Phase III was the demonstration test of the complete fuel cell energy recovery sys-
tem. Phase IV described how the commercial fuel cell power plant could be further
modified to achieve full rated power on LHV landfill gas. The demonstration test
successfully demonstrated operation of the energy recovery system, including the
GPU and the commercial phosphoric acid fuel cell modified for operation on landfill
gas. Demonstration output included operation up to 137 kW; 37.1% efficiency at 120
kW; exceptionally low secondary emissions (dry gas, 15% O2) of 0.77 ppmV carbon
monoxide, 0.12 ppmV nitrogen oxides, and undetectable sulfur dioxide; no forced out-
ages with adjusted availability of 98. 5%; and 709 hours operation on landfill gas.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. coSATl Field/Group
Pollution
Energy
Fuel Cells
Phosphoric Acids
Earth Fills
Gases
Methane
Carbon Dioxide
Sulfur
Halides
Pollution Prevention
Stationary Sources
Global Warming
13 B
14G
10B
07B
13 C
07D
07C
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
88
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 13-73)
78
------- |