&EPA
united States
Environmental Protection
Agency
Industrial Environmental Research EPA-600/7-79-105b
Laboratory April 1979
Research Triangle Park NC 27711
Comparative Assessment
of Residential Energy
Supply Systems That
Use Fuel Cells
(Technical Report)
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide'range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-79-105b
April 1979
Comparative Assessment of Residential
Energy Supply Systems That Use Fuel Cells
(Technical Report)
by
,R.V. Steele, D.C. Bomberger, K.M. Clark, R.F. Goldstein, R.L Hays, M.E. Gray
and
G. Ciprios, R.J. Bellows, H.H. Horowitz, C.W. Snyder (Exxon)
SRI International
333 Ravenswood Avenue
Menlo Park, California 94025
Contract No. 68-02-2180
Program Element No. EHB534
EPA Project Officer: Gary L Johnson
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
SRI INTERNATIONAL
COMPARATIVE ASSESSMENT OF RESIDENTIAL
ENERGY SUPPLY SYSTEMS THAT USE FUEL CELLS
EXECUTIVE SUMMARY
What Are Fuel Cells?
Fuel cells are devices capable of converting the
chemical energy stored in a fuel directly into
electrical energy without a step involving combustion.
Hydrogen contained in the fuel is chemically combined
with oxygen from the air to produce water and an
electric current that can be regulated and used.
Fundamentally, the process is just the inverse of the
electrolysis of water into its component parts, a
process often demonstrated in high school chemistry
classes. Practically, a fuel cell consists of two
electrodes, a catalyst used to promote the chemical
reaction, and an electrolyte (a chemical substance that
conducts electricity) separating the electrodes. As
might be suspected, a device of such fundamental
simplicity was first conceived long ago—in 1839 by Sir
William Grove, a British jurist.
Are Fuel Cells Commercially
Available Today?
Although old in concept, as practical devices for
producing electricity in significant amounts, fuel cells
are in their infancy. For space missions, fuel cells
have been shown to be ideal power sources, partly
because they convert on-board stores of hydrogen and
oxygen to electrical power without producing
excessive heat or vibration-producing mechanical
motion. In fact, they provided electrical power in
Gemini and Apollo spacecraft, but were still
considered novel and exotic devices.
-------
Made in limited quantities, and to extreme reliability
standards, fuel cells for space craft are understandably
expensive. Nevertheless, much has been learned from
the space program about fuel cells and that knowledge
is beginning to find earthbound applications in
much-improved and less costly devices.
More than 60 small (12.5 kW) fuel-cell power plants
were field tested in 1972 and 1973. A 40-kW device
was demonstrated in 1975, and now work is underway
to demonstrate a 4.5-MW fuel cell in the Consolidated
Edison (New York) utility system by 1980. Fuel-cell
technology has come a long way and is near ing
commercial readiness.
Do Fuel Cells Possess Much of the present interest in fuel cells derives from
Attractive Attributes? their unusually low environmental impact and their
high efficiency. Because no combustion is involved,
even fuel cells that use common fuels produce very
low emissions of nitrogen or sulfur oxides; the
emissions are many times below federal standards.
Moreover, fuel cells generally consume no water and
operate very quietly.
As a result of its environmental good- neighborliness, a
fuel-cell power plant can easily be located very near
the power demands it serves, thereby lessening the
need for high voltage electric transmission lines.
-------
The ability to site fuel-cell power plants locally is
much enhanced by their modular design (which allows
off-site manufacturing) and their rapid installation.
Accordingly, electric power Utilities may soon have
commercially available a device that enables system
expansion in small increments.
How Might Fuel Cells Fit
into Electric Power Systems?
Besides being suitable for small, dispersed, locally
sited power stations, fuel cells can easily operate in
applications that require output to follow demand
closely. In fact, electric utility interest in fuel cells
often centers on mid-1980s deployment for
load-following. Again, because of their cleanliness,
fuel cells may be installed in buildings or residential
complexes where the combined production of electric
power and heat could be used to satisfy heating and
cooling demands in an integrated (or "cogeneration")
fashion. The fuel-cell-derived electricity would be
used to operate heat pumps to provide cooling and
supplemental heating.
Can Fuel Cells Use Coal
or Coal-Derived Fuels?
Fuel cells, like most fuel-consuming devices are
indifferent to the origin of the fuel—as long as in final
form it conforms to the chemical requirements of the
device. Accordingly, natural gas, petroleum products,
or similar fuels are perfectly acceptable in fuel cells
provided that the fuels are first reformed to hydrogen
and carbon dioxide and that harmful sulfur
contamination is removed before the fuels enter the
fuel cell proper.
iii
-------
What Are Leading Coal-Based
Alternatives to Fuel Cells?
Fuel cells, therefore, can have a place in a largely
coal-based U.S. energy future.
Already, U.S. electrical power is largely generated in
coal-fired plants and the federal government is pushing
for even more in an effort to save relatively scarce
and expensive oil and natural gas for other uses.
Larger, conventional coal-fired power plants, often
located in remote areas and connected to urban load
centers by high voltage transmission lines, certainly
provide a well-proven alternative to electric power
generated from fuel cells.
So-called "combined-cycle" electrical power
generation—a conventional boiler and steam turbine
generator supplemented by a high-temperature gas
turbine—is an improving technology gaining
considerable attention among utilities. Certainly, by
the time fuel-cell systems are perfected sufficiently
to allow commercial deployment, combined-cycle
systems will already be in use and fuel-cell systems
will have to compete with them.
Much of the U.S. space heating demand is met by the
combustion of natural gas. Because so many
consumer-owned heaters are already in place, gas
utilities have strong incentive to supplement natural
gas supplies with coal-derived substitutes that would
not require alteration of either consumer appliances or
habits.
IV
-------
Can Fuel Cells be Compared
with the Alternative
Technologies?
Synthetic natural gas, (SNG) derived from coal, then,
offers strong competition for the electric heating role
a fuel-cell/heat pump combination could play in the
market place.
Because fuel cells must compete with so many electric
power and heat-producing fuel and technology
combinations, the relative advantages and
disadvantages of fuel cells have proven difficult to
discern clearly. Consequently, as a part of its mission
to preserve and enchance environmental quality, the
U.S. Environmental Protection Agency commissioned
this study precisely to learn more about what might be
expected from fuel cells when actually deployed in
utility systems.
To address this question, SRI International
conceptually designed twelve energy systems able to
provide residential heating and cooling using
technologies projected to be available toward the end
of this century. Only a few systems used fuel cells.
As in most comparisons, some constraints were
imposed to eliminate unnecessarily confusing
complexities while providing a uniform framework for
comparison. Accordingly, all systems use western coal
as the primary energy resource, and all residences are
assumed to have identical heating and cooling demands
typical of the mid-continent United States. After
winnowing out the clearly least attractive
combinations, we selected five systems and compared
them in great detail.
-------
For all the comparisons, we examined the entire chain
of the system, starting with the coal mine and ending
with the heating and cooling of residences, to be sure
that the claimed environmental advantages of the fuel
cells at the point of electric power generation did not
distract us from some important environmental
impacts elsewhere in the system. Our five surviving
systems, four of which use heat pumps for heating and
cooling are:
o System 1—A coal-fired power plant supplies
electricity and a coal gasification plant supplies
SNG to residences; electricity powers air
conditioners and SNG is burned in gas furnaces.
o System 2—A 26-MW fuel-cell power plant fueled
by coal-derived SNG supplies electricity to
residences with heat pumps.
o System 3—A 26-MW fuel-cell power plant fueled
by coal-derived naphtha supplies electricity to
residences with heat pumps.
o System 4—A combined-cycle power plant fueled
by coal-derived fuel oil supplies electricity to
residences with heat pumps.
o System 5—A 100-kW fuel-cell power plant fueled
by coal-derived SNG, sited in a housing complex,
supplies electricity to townhouses with heat
pumps; heat recovered from the fuel cell supplies
supplemental space heating and hot water.
Of these five, the first one most resembles the
existing order in the utility industry, and the fourth
constitutes an already evident evolutionary change of
the industry.
VI
-------
What do the Comparisons
Show?
The scorecard for the various systems is mixed—no
single system stands out as superior in all the
attributes that will ultimately decide which systems
will be deployed. Nevertheless, some very interesting
facts emerge about energy systems that use fuel cells.
Which System Costs the
Consumer More?
The three fuel-cell systems provide heating and
cooling to our standard residences at considerably
higher cost that the two more conventional systems.
In fact, the annual energy bill to a consumer using
System 5 is over 63% higher than for one using System
1, the most conventional and lowest cost option. The
order of cost, from the least expensive system to the
most expensive, is 1,4,2,3,5.
Are There Differences in the
Capital Investment Required?
Which Has the Best System
Performance?
The scorecard for the capital intensiveness of the five
systems largely follows the pattern of the annual cost
to consumers. In order, from least to most capital
intensive, are Systems 1, 4, 3, 2, 5. Because capital is
itself a scarce resource, utilities most likely will show
most interest in Systems 1 and 4.
Because all five systems contain at least one element
not yet proven in commercial service, such things as
reliability, the degree of redundancy needed in a
system, and the ability to integrate smoothly the new
devices into a system are difficult to assess, more so
than for most other comparison attributes. We judge,
however that, overall, the most conventional system is
most likely to give the best performance. System
performance, from best to worst comes in this order:
Systems 1, 2, 4, 3, 5.
Vll
-------
Which System is Most
Efficient?
When making a comparison of system efficiency, we
were careful to account for energy losses at every step
in proceeding from the coal mine to the heated and
cooled residence. All fuel-cell systems are
considerably more efficient than the most
conventional system, System 1. Indeed, System 5 is
75% efficient, while System 1 is only 41% efficient.
Systems 2, 3, and 4 possess nearly equal efficiencies in
the 64% to 67% range. This attribute is particularly
important because it shows that the systems using fuel
cells required less coal to accomplish the same end—a
virtue that, besides conserving resources, carries over
into lessened environmental impact.
What About Air Quality?
Are There Differences in
Water Quality?
Because maintenance of air quality around electric
power generation plants is a vexing and costly
problem, the relative scores for this indicator could
prove especially important to utilities in the years
ahead. We weighted equally pollutants emitted at the
fuel production site and the fuel consumption site
(both overwhelm the emissions from fuel
transportation). Again, all three systems using fuel
cells are superior to the two more conventional
systems, with System 5 being the cleanest and System
1 emitting the most pollutants. In order, from least to
most polluting are Systems 5, 2, 3, 4, 1.
For this indicator we weighted equally effluents and
water consumption at the fuel production and the fuel
consumption locations.
viii
-------
All three fuel-cell systems are cleaner than the two
more conventional systems. Again, System 5 is the
cleanest, but this time System 4 degrades water
quality the most. In order of cleanliness are Systems
by £y <5 y lj Tt«
How Do They Compare on
Solid Waste?
Most solid waste for this set of five systems is
produced as ash in converting the coal to a more useful
energy form. Consequently, scores in this category
essentially mirror the overall system energy efficiency
ratings—the most efficient System 5 also produces the
least solid waste and the least efficient System 1
produces the most solid waste. Systems 2, 3, 4 are
nearly tied, and produce about the same intermediate
quantities.
What About Land Use, Noise
and Aesthetics?
The three parameters are closely linked because
aesthetics and human exposure to noise produced are
greatly affected by location and the amount of land
occupied or disturbed. Overall, least obtrusive is
System 5 and the most obtrusive is System 1.
Is There a Pattern in the
Comparison?
A striking pattern emerges when we assemble the
scores for all categories of comparison. The fuel-cell
systems are the most costly—to build and install as
well as in end-use cost to consumers—but are the most
environmentally benign and consume the least coal to
get the heating and cooling job done.
ix
-------
We expected from the outset of this study that the
fuel cells themselves would be clean compared to
alternatives, but our finding that entire fuel-cell
systems from resource extraction to final demand
offer overall environmental benefits is new.
Will Fuel Cell Systems
Actually Be Used?
How the trade-off between environmental cleanliness
and economic cost will be valued in the next several
decades will prove crucial to the question of whether
fuel-cell systems resembling those we have examined
will actually be deployed in meaningful numbers. One
thing is certain: Fuel-cell systems possess a mixture
of attributes much different from the more
conventional electric power systems. As a result, U.S.
utilities will have available an important new electric
power option in the years ahead.
Full analysis is available in the 500-page report:
"Comparative Assessment of Residential Energy
Supply Systems That use Fuel Cells," Environmental
Protection Agency, Report No. 600/7-79-105b, 1979.
-------
ACKNOWLEDGMENTS
The work reported here was sponsored by the U.S. Environmental
Protection Agency under Contract 68-02-2180. The EPA Project officer
was Gary L. Johnson of the Special Studies Staff, Industrial Environ-
mental Research Laboratory, Research Triangle Park, North Carolina.
The bulk of the project work was carried out at SRI International,
Menlo Park, California, under the leadership of Dr. Robert V. Steele in
the Center for Resource and Environmental Systems Studies. Overall
project supervision was provided by Dr. Edward M. Dickson who also wrote
the executive summary and aided in compiling and editing the final
report.
Most of the systems analysis, and the analysis of residential
heating and cooling requirements, was carried out by Dr. Steele. Roger
Goldstein, assisted by other members of SRI's Energy Center, carried out
the engineering and economic analysis of energy system components (ex-
cluding fuel cells and residential heating and cooling equipment).
Specification of environmental control requirements for coal conversion
facilities was done by Dr. David Bomberger of the Environmental Control
Department while environmental analysis of other energy system com-
ponents was performed by Roy Hays, Mary Gray and Kristin Clark.
Under subcontract to SRI International, Exxon Research and
Engineering Co., Lindon, New Jersey, carried out all the conceptual
design work, cost and efficiency calculations, and environmental analy-
sis for the phosphoric acid and molten carbonate fuel-cell power
plants. Exxon also provided the overview of fuel-cell technology
XI
-------
presented in Chapter II. George Ciprios of Exxon supervised the sub-
contracting effort.
Others who assisted in the report but not cited as authors include
Dr. Buford R. Holt who assisted in the compilation of environmental
impact data and in the development of criteria for assessing the re-
lative impacts of air pollutants. The procedure for calculating
residential heating and cooling loads was programmed by Dr. Barry
Scott-Walton. The editor for the final report was Barbara J. Stevens,
and the illustrations were done by Grace Tsai and L. H. Wu.
xii
-------
CONTENTS
LIST OF ILLUSTRATIONS xix
LIST OF TABLES xxv
GLOSSARY OF ABBREVIATED TERMS xxxiii
GLOSSARY OF UNITS xxxiii
I INTRODUCTION 1-1
II OVERVIEW OF FUEL-CELL TECHNOLOGY II-l
A. The Fuel-Cell Concept II-l
B. Fuel Cell Applications II-4
C. Potential Fuel Cell Advantages and Disadvantages . . II-7
D. The Complete Fuel-Cell System 11-11
1. Fuel-Cell Reactants 11-12
2. Fuel-Cell Operating Characteristics 11-16
3. Fuel-Cell Economics 11-20
4. Fuel-Cell Technologies and Trade-Offs 11-24
E. Specific Fuel-Cell Technologies 11-26
1. Phosphoric Acid Fuel Cells 11-26
2. Molten Carbonate Fuel Cells 11-29
3. Solid Oxide Fuel Cells 11-34
4. Alkaline Fuel Cells 11-36
5. Ion Exchange Membrane Cells 11-38
III SELECTION OF ENERGY SUPPLY SYSTEMS FOR DETAILED ANALYSIS. . . III-l
A. Criteria for Proposed Systems III-l
B. Proposed Systems III-3
1. Type 1: Conventional Power Plant/SNG III-3
.2. Type 2: 26-MW Fuel-Cell Power Plant Ill-8
3. Type 3: Combined-Cycle Power Plant 111-10
4. Type 4: 100-kW Fuel-Cell Power Plant
with Heat Recovery 111-10
Xlll
-------
C. Evaluation of System Components 111-13
1. Cost and Efficiency 111-13
2. Sulfur Dioxide Emissions 111-14
3. Ranking of Proposed Systems I11-16
D. Selection of Systems 111-16
E. References—Chapter III 111-22
IV CONCEPTUAL SYSTEMS DESIGNS IV-1
A. System 1 IV-2
1. Coal Mine IV-2
2. Unit Train IV-7
3. Coal-Fired Power Plant IV-10
4. Electricity Transmission and Distribution . . . IV-13
5. Coal Gasification Facility IV-16
6. Gas Pipeline IV-24
7. Gas Distribution IV-25
8. Gas Furnace and Air Conditioner IV-26
B. System 2 IV-29
1. Coal Mine IV-29
2. Coal Gasification Facility IV-29
3. Gas Pipeline IV-29
4. Gas Distribution IV-29
5. 26-MW Fuel-Cell Power Plant IV-31
6. Distribution of Electricity IV-47
7. Heat Pump IV-48
C. System 3 IV-53
1. Coal Mine IV-53
2. Coal Liquefaction Plant IV-53
3. Liquids Pipeline IV-61
4. Distribution of Naphtha IV-62
5. 26-MW Fuel-Cell Power Plant IV-63
6. Distribution of Electricity IV-72
7. Heat Pump IV-72
D. System 4 IV-74
1. Coal Mine IV-74
2. Coal Liquefaction Plant IV-74
3. Liquids Pipeline IV-78
4. Fuel Distribution IV-78
5. Combined-Cycle Power Plant IV-79
6. Transmission and Distribution
of Electricity IV-82
7. Heat Pump IV-82
E. System 5 IV-84
1. Coal Mine IV-84
xiv
-------
2. Coal Gasification Facility IV-84
3. Gas Pipeline IV-84
4. Gas Distribution IV-84
5. 100-kW Fuel-Cell Power Plant IV-84
6. Heating and Cooling System IV-104
F. References—Chapter IV IV-112
V THERMAL EFFICIENCY OF THE SYSTEM COMPONENTS V-l
A. Coal Mine V-2
B. Unit Train V-3
C. Coal-Fired Power Plant V-4
D. Coal Gasification Plant V-5
E. Coal Liquefaction Plant V-7
F. Gas Pipeline V-10
G. Liquids Pipeline V-10
H. Liquid Fuel Distribution V-ll
I. Gas Distribution V-12
J. Combined-Cycle Power Plant V-12
K. 26-MW Fuel-Cell Power Plant (SNG) V-14
L. 26-MW Fuel-Cell Power Plant (Naphtha) V-15
M. Electricity Transmission and Distribution V-15
N. 100-kW Fuel-Cell Power Plant
with Heat Recovery V-l7
0. Gas Furnace and Air Conditioner V-21
P. Heat Pumps V-23
Q. Heat Delivery System V-26
R. References—Chapter V V-30
VI ENVIRONMENTAL IMPACTS OF SYSTEM COMPONENTS VI-1
A. Coal Mine VI-1
xv
-------
1. Environmental Setting VI-2
2. Surface Mining Operation Requirements VI-3
3. Environmental Impacts of Surface Mining .... VI-5
B. Unit Train (Coal) VI-10
1. Loading/Unloading VI-10
2. Line Haul VI-10
C. Coal-Fired Power Plant VI-15
1. Flue Gas Treatment System VI-18
2. Solids Handling System VI-20
3. Pollutant Emissions VI-21
D. Coal Gasification Plant VI-28
1. Sulfur Recovery Plant VI-32
2. Process Condensate Treatment System VI-34
3. Solids Processing system VI-37
4. Pollutant Emissions VI-37
E. Coal Liquefaction Plant VI-42
1. Pollution Control System VI-42
2. Pollutant Emissions VI-47
F. Pipelines VI-50
1. Physiography VI-50
2. Hydrology VI-52
3. Vegetation VI-53
4. Wildlife VI-54
5. Air Quality VI-55
G. Fuels Distribution VI-56
H. Combined-Cycle Power Plant VI-61
I. 26-MW Fuel-Cell Power Plant (SNG) VI-64
1. Spent Fuel-Cell Stack Disposal VI-66
2. Spent Catalyst Guard Bed Disposal VI-66
3. Noise VI-67
J. 26-MW Fuel-Cell Power Plant (Naphtha) VI-67
K. Electricity Transmission and Distribution VI-69
1. Transmission Line Characteristics VI-70
2. Transmission Line Impacts VI-71
L. Gas Furnace . VI-77
M. 100-kW Fuel-Cell Power Plant VI-77
1. Spent Fuel-Cell Stack Disposal VI-81
2. Spent Catalyst Bed Disposal VI-81
3. Normal Power Plant Operation VI-82
0. References—Chapter VI VI-83
xv i
-------
VII CAPITAL
A.
B.
C.
D.
E.
F.
G.
H.
I.
J.
K.
L.
M.
N.
0.
P.
Q.
R.
VIII THERMAL
A.
B.
C.
AND OPERATING COSTS OF SYSTEM COMPONENTS
Unit Train
26-MW Fuel-Cell Power Plant (SNG)
1. Fuel-Cell Trailer Cost
6. Total Power Plant
26-MW Fuel-Cell Power Plant (Naphtha)
100-kW Fuel-Cell Power Plant
AND ELECTRICAL LOAD FACTORS
Thermal Response of the Residences
. . VII-1
. . VII-3
, , VII-7
VII-7
, , VII-13
. . VII-13
. , VII-20
. . VII-27
. . VII-32
VII-35
. . VII-38
. . VII-38
. . VII-42
. . VI 1-46
. . VII-46
. . VII-47
. . VII-47
. . VII-48
. . VII-53
. . VII-62
. . VII-66
. . VII-7 2
. . VII-76
. . VII-78
. . VII-79
. . VIII-1
. . VIII-2
. . VIII-5
. . VIII-7
KVll
-------
D. Energy Consumption for Heating and Cooling VIII-12
E. References—Chapter VIII VIII-24
IX COMPARATIVE ANALYSIS IX-1
A. Energy Efficiency IX~1
B. Economics IX-10
1. Cost of Heating and Cooling IX-10
2. Capital Intensiveness IX-19
C. Environmental Impact IX-24
1. Pollutant Emissions IX-24
2. Land Use, Noise, and Aesthetics IX-37
D. System Performance IX-42
E. References—Chapter IX IX-48
X SUMMARY AND CONCLUSIONS X-l
A. Summary of Advantages and Disadvantages X-l
1. System 1 X-l
2. System 2 X-3
3. System 3 X-3
4. System 4 X-3
5. System 5 X-3
B. Conclusions X-4
APPENDICES
A ENTHALPIES BASED ON THE
GIRDLER CATALYSTS DATA HANDBOOK A-l
B MOLTEN CARBONATE FUEL CELL PERFORMANCE B-l
C ENERGY SUPPLY/DEMAND PROGRAM FOR RESIDENCES
SUPPLIED BY THE 100-kW FUEL-CELL POWER PLANT C-l
xviii
-------
LIST OF ILLUSTRATIONS
II-l The Fuel-Cell Concept H-2
II-2 Conventional Energy Generation II-2
II-3 Efficiency Characteristics of Total Energy System. . . . II-5
II-4 Efficiency of Various Energy Conversion Devices II-8
II-5 Part-Load Efficiency II-8
II-6 The Fuel-Cell System 11-12
II-7 The EMF Series 11-14
II-8 Fuel-Cell Losses 11-18
II-9 Characteristic Fuel-Cell Performance Curve 11-18
11-10 Fuel-Cell Power Plant 11-19
11-11 Effect of Current Load on Voltage Efficiency 11-19
11-12 Capital Charges 11-21
11-13 Fuel Charges 11-21
11-14 Stack Replacement Charges 11-23
11-15 Design Point Selection 11-23
11-16 Progress in Molten Carbonate Fuel-Cell Performance . . . 11-32
III-l Type la: Coal-Fired Power Plant/SNG III-5
III-2 Type Ib: Oil-Fired Power Plant/SNG III-7
III-3 Type 2: 26-MW Fuel-Cell III-9
III-4 Type 3: Combined-Cycle Power Plant III-l 1
III-5 Type 4: 100-kW Fuel Cell with Heat Recovery 111-12
IV-1 Block Flow Diagram of System 1 IV-3
IV-2 Dragline Method of Overburden Removal IV-5
IV-3 Block Flow Diagram for an 800-MW Coal-Fired Power Plant. IV-12
xix
-------
IV-4 High-Voltage Transmission Economy for Long Distance. . . IV-14
IV-5 Block Flow Diagram of Hygas SNG Plant IV-19
IV-6 Cooling Capacity of the Westinghouse SL030C/EC030
Air Conditioner IV-28
IV-7 Block Flow Diagram of System 2 IV-30
IV-8 Block Flow Diagram for 26-MW Fuel-Cell Power Plant
(SNG Fuel) IV~33
IV-9 Fuel Cell Trailer Layout IV~40
IV-10 Reformer/Heat Exchanger Package IV-43
IV-11 Equipment Module Layout IV-45
IV-12 Power Plant Layout IV-46
IV-13 Recommended Heat Pump Configuration for Split System
Air-to-Air Units IV-50
IV-14 Heating and Cooling Capacity of 26.0 MJ/hr
(24,600 Btu/hr) Heat Pump IV-52
IV-15 Block Flow Diagram of System 3 IV-54
IV-16 H-Coal Process Flow Diagram IV-56
IV-17 Block Flow Diagram for 26-MW Fuel-Cell Power Plant
(Naphtha Fuel) IV-64
IV-18 Block Flow Diagram of System 4 IV-75
IV-19 H-Coal Process Flow Diagram IV-76
IV-20 Combined-Cycle Power Plant Block Flow Diagram IV-81
IV-21 Block Flow Diagram of System 5 IV-85
IV-22 100-kW Power Plant Components IV-87
IV-23 Heat Integration for 100-kW Fuel-Cell Power Plant. . . . IV-88
IV-24 Phosphoric Acid Fuel Cell Performance Data IV-93
IV-25 Phosphoric Acid Fuel Cell Performance Curve IV-97
IV-26 Reformer Package for 100-kW Power Plant IV-102
IV-27 System Layout for 100-kW Power Plant IV-105
xx
-------
IV-28 Site Plan for 100-kW Fuel-Cell Power Plant and Townhouses IV-107
IV-29 Schematic of Hot Water and Space Heat System Using
Recovered Fuel-Cell Heat IV-108
IV-29 Heating and Cooling Capacities of 19.3 MJ/hr
(18,300 Btu/hr) Heat Pump IV-110
V-l Effect on Thermal Efficiency of Operating the 100-kW Total
Energy System at Part Load V-19
V-2 Heat Availability for 100-kW Fuel-Cell Power Plant . . . V-20
V-3 Effect of Water Return Temperature on Overall
Thermal Efficiency. .... V-22
V-4 Coefficient of Performance of Westinghouse SL030
Air Conditioner V-24
V-5 Coefficients of Performance of Advanced Heat Pumps . . . V-27
V-6 Heat Transfer Rates From Hot Water Delivery System
to Space Heating V-29
VI-1 Integrated Pollution Control System for an 800-MW
Coal-Fired Power Plant VI-17
VI-2 Solids Handling System for an 800-MW
Coal-Fired Power Plant VI-22
VI-3 Integrated Air, Water, and Solids Pollution Control
System for a Hygas Coal Gasification Plant
That Produces 7.8 Million nm3 of SNG per Day . . . VI-31
VI-4 Wastewater Treatment Plant for Coal Gasification .... VI-36
VI-5 Integrated Pollution Control System for an H-Coal
Liquefaction Plant That Produces 7,950 nm3
per Day of Fuel VI-46
VII-1 Sensitivity of Unit Train Costs to Volume of Traffic
on New Portion (80 km) of Track VII-10
VII-2 Sensitivity of the Cost of Electricity to Plant Capital
Cost and Delivered Coal Cost VII-14
VII-3 Sensitivity of the Cost of SNG to Plant Capital Cost
and Coal Cost VII-17
VII-4 Sensitivity of the Cost of Distillate Fuel Oil to
Plant Capital Cost and Coal Cost VII-23
VII-5 Sensitivity of the Cost of Hydrotreated Naphtha to
Plant Capital Cost and By-Product Fuel Oil Credit . VII-24
xx i
-------
VII-6
VII-7
VI1-8
VII-9
VII-10
VII-11
VII-12
VII-13
VII-14
VIII-1
VIII-2
VIII-3
VIII-4
VIII-5
VIII-6
IX-1
IX-2
IX-3
IX-4
Sensitivity of SNG Transmission Cost to Pipeline
Capital Cost and SNG Cost
Effect of Pipe Diameter on SNG Transmission Costs.
Sensitivity of Liquid Fuel Transmission Cost to
Pipeline Capital Cost
Railroad Tank Car Transport Costs (Fuel Oil)
Tank Truck Transportation Costs (Naphtha). .
Sensitivity of Cost of Electricity to Power Plant
Capital Cost and Distillate Fuel Cost . . . .
Sensitivity of the Cost of Electricity to Power Plant
Capital Cost and SNG Cost
Sensitivity of the Cost of Electricity to Power Plant
Cost and Naphtha Cost
Sensitivity of the Cost of Electricity to Load Factor
and Hot Water Credit
Variation in Hourly Average Light and Appliance Loads
and DHW Demand with Time of Day - Residence 3 . .
DHW Demand Relative to Fuel Cell Heat Available from
Operations of Lights and Appliances
Hourly Average Temperature, Heating Load, and Supply
of Recovered Fuel Cell Heat for Residence 3 on the
Coldest Day of the Year
Hourly Average Electrical Loads for Residence 3 on the
Coldest Day of the Year
Variations in Average Space Heating Electrical Load
and Delivered Fuel Heat with External Temperature .
Hourly Average Temperature and Electrical Load for
Residence 3 on the Hottest Day of the Year
Annual Energy Flows and Energy Efficiency
for System 1
Annual Energy Flows and Energy Efficiency
for System 2
Annual Energy Flows and Energy Efficiency
for System 3
Annual Energy Flows and Energy Efficiency
for System 4
VII-28
VII-29
VII-33
VII-34
VII-36
VII-41
VII-54
VII-61
VII-74
VIII-4
VIII-16
VIII-17
VIII-18
VIII-20
VIII-22
IX-4
IX-5
IX-6
IX-7
xxii
-------
IX-5 Annual Energy Flows and Energy Efficiency
for System 5 IX-8
IX-6 Total Annual Energy Flows (per Residence)
for Fuel-Cell Power Plant Supplying Townhouses. . . IX-9
IX-7 Variation in the Cost of Heating and Cooling
with Heating Load - System 5 IX-16
C-l Program for Energy Supply/Demand Calculations C-l
xxiii
-------
LIST OF TABLES
II-l Efficiency Limits of Conventional Thermal
Energy Conversion Systems II-3
11-2 Full Selection Trade-Off 11-15
II-3 Estimated Installed Investment Costs
for Fuel-Cell Power Plants 11-24
II-4 Fuel-Cell Technology Classifications 11-25
III-l Categorization of Proposed Residential
Energy Supply Systems III-4
III-2 Ranking of Cost, Efficiency, and SC^ Emissions
for Energy Supply Systems III-l7
IV-1 Main Process Flows for an 800-MW
Coal-Fired Power Plant IV-14
IV-2 Economic Power Loading of Transmission Lines IV-15
IV-3 Main Process Flows for SNG from Coal
via the Hygas Process IV-22
IV-4 Process Flow Streams for 26-MW Fuel-Cell
Power Plant (Naphtha) IV-37
IV-5 Equipment List for 26-MW Fuel-Cell Power Plant (SNG) . . IV-38
IV-6 Piping Specifications IV-47
IV-7 Composition of Major Streams in H-Coal Process IV-57
IV-8 Properties of H-Coal Distillate Fuel Oils IV-58
IV-9 Process Flow Steams for 26-MW Fuel-Cell
Power Plant (Naphtha) IV-69
IV-10 Equipment List for 26-MW Fuel-Cell
Power Plant (Naphtha) IV-73
IV-11 Composition of Major Streams in H-Coal Process IV-77
IV-12 Mass Flow Rates for 270-MW
Combined-Cycle Power Plant IV-83
IV-13 Additional Details for Figure IV-23 IV-94
xxv
-------
IV-14 Process Conditions and Flow Rates
for 100-kW Fuel-Cell Power Plant .......... IV-98
IV-15 100-kW Fuel-Cell Power Plant .............. IV-101
V-l Energy Balance for an 800-MW Coal-Fired Power Plant
That Uses Subbituminous Coal ............ v~5
V-2 Energy Balance for a 7.8 x 10^ urn3
(275 x 106scf) per Day Coal Gasification
Plant Based on the Hygas Process .......... v~6
V-3 Energy Balance for a 7950 m3 (50,000 bbl)
per Day Coal Liquefaction Plant That
Produces Distillate Fuel Oil ............ V-8
V-4 Energy Balance for a 7630 m3 (48,000 bbl)
per Day Coal Liquefaction Plant That
Produces Naphtha and Fuel Oil ........... V-9
V-5 Energy Balance for a 270-MW
Combined-Cycle Power Plant ............. V-12
V-6 Estimated Heat Rates at Part Load for Combined-Cycle
Power Plant .................... V-13
V-7 Energy Balance for a 24.0-MW Fuel-Cell
Power Plant that Uses SNG ............. V-14
V-8 UTC Projection of Part-Load Heat Rate
for Molten Carbonate Fuel Cells
Using Reformable Fuels ............... V-16
V-9 Energy Balance for a 25.6-MW Fuel-Cell
Power Plant That Uses Naphtha ........... V-16
V-10 Energy Balance for a 100-kW Fuel-Cell
Power Plant That Uses SNG ............. V-18
V-ll Operating Characteristics of 100-kW Total Energy
Power Plant at Part Load .............. V-18
VI-1 Projected Particulate Emissions from Surface
Mines — Gillette Area ................ VI-6
VI-2 Air Pollutant Emissions for a 4.5 Million Tonne
(5 Million Ton) per Year Surface Mine ....... VI-8
VI-3 Air Pollutant Emissions from Diesel Locomotives —
100-Car Unit Coal Train .............. VI-14
VI-4 Emissions from a Coal-Fired Power Plant ......... VI-16
xxv i
-------
VI-5 Summary of Major Emissions from
an 800-MW Coal-Fired Power Plant .......... VI-23
VI-6 Concentrations of several Toxic Trace Elements
in Powder River Basin Coal ............. VI-26
VI-7 Emission of Toxic Trace Elements from an 800-MW
Coal-Fired Power Plant ............... VI-27
VI-8 Emissions of Polycyclic Aromatic Hydrocarbons
from an 800-MW Coal-Fired Power Plant ....... VI-28
VI-9 Major Sources of Pollution
for a Coal Gasification Plant ........... VI-30
VI-10 Major Emissions from Hygas Coal Gasification
(7.8 Million nm3 of SNG per Day) ......... VI-39
VI-11 Combustion-Related Emissions of Fine Particulates,
Trace Elements, and PAH from the Hygas Coal
Gasification Plant .............. . . VI-40
VI-12 Major Sources of Pollution
for a Coal Liquefaction Plant ........... VI-43
VI-13 Major Emissions from H-Coal Liquefaction
(7,950 m3 of Distillate Fuel Oil per Day) .... VI-48
VI-14 Combustion-Related Emissions of Fine Particulates,
Trace Elements, and PAH from an
H-Coal Liquefaction Plant ............. VI-49
VI-15 Air Pollutant Emissions from a Compressor Station
on an 81 cm (32 in.) Natural Gas Pipeline ..... VI-57
VI-16 Air Pollutant Emissions from a Pumping Station
on a 51 cm (20 in.) Liquid Fuels Pipeline ..... VI-57
VI-17 Diesel Truck Noise Sources ............... VI-59
VI-18 Urban and Suburban Detached Housing Residential
Areas and Approximate Daytime Residual
Noise Level (L) ................ VI-60
VI-19 Air Pollutant Emissions from
a Diesel-Powered Tank Truck ............ VI-61
VI-20 Emissions from a 270-MW Combined-Cycle Power Plant . . . VI-62
VI-21 Emissions of Toxic Trace Elements from a 270-MW
Combined-Cycle Power Plant ............. VI-63
VI-22 Emission of PAH from a 270-MW
Combined-Cycle Power Plant ............. VI-64
xxv ii
-------
VI-23 Composition of Effluent Process Stream
for a 24.0-MW Fuel-Cell Power Plant VI-65
VI-24 Predicted Reformer Furnace NOX Production VI-65
VI-25 Composition of Effluent Process Stream
for a 25.6-MW Fuel-Cell Power Plant VI-68
VI-26 Predicted Reformer Furnace NOX Production VI-68
VI-27 Average Requirements for Rights-of-Way VI-71
VI-28 60-Hz Electric and Magnetic Fields VI-75
VI-29 Emission of Air Pollutants from a
70-MJ/hr Residential Gas Furnace VI-78
VI-30 Composition of Effluent Stream from
a 100-kW Fuel-Cell Power Plant VI-78
VI-31 Predicted Reformer Furnace NOX Production VI-79
VI-32 Published Pollution Characteristics of Experimental
Phosphoric Acid Fuel Cells VI-80
VII-1 Capital Investment Required for a 4.5 Million
Tonne (5 Million Ton) per Year Surface Coal
Mine in the Powder River Basin VII-5
VII-2 Capital Investment Required for a 4.5 Million
Tonne (5 Million Ton) per Year Surface Coal
Mine in the Powder River Basin VII-6
VII-3 Estimated Investment for 80 km (50 mi) of New Track
and Two 100-Car Unit Trains VII-8
VII-4 Operating Cost and Revenue Required for New
Unit Train, 80 km (50 mi) New Track, and
1,200 km (750 mi) Existing Track VII-9
VII-5 Capital Investment for 800-MW
Coal-Fired Power Plant with FGD VII-11
VII-6 Operating Costs and Revenue Requirements
for 800-MW Coal-Fired Power Plant with
FGD (35% Load Factor) VII-12
VII-7 Investment Required for a 7.8 x 10^ m^
(275 x lO^scf) per Day SNG Plant Based
on the Hygas Process VII-15
VII-8 Operating Costs and Revenue Requirements for a
7.8 x 106 nm3 (275 x 106 scf)
per Day SNG Plant Based on the Hygas Process. . . . VII-16
xxviii
-------
VII-9 Capital Investment for a 7,950 nm3 (50,000 bbl)
per Day Plant Producing Distillate Fuel Oil
from Coal by the H-Coal Process VII-18
VII-10 Capital Investment for 7,630 nm3 (48,000 bbl)
per Day Plant Producing Naphtha and Fuel Oil
from Coal by the H-Coal Process VII-19
VII-11 Operating Costs and Revenue Requirements for
a 7,950 nm3 (50,000 bbl) per Day Plant
Producing Distillate Fuel Oil from Coal
by the H-Coal Process VII-21
VII-12 Operating Costs and Revenue Requirements for
a 7,630 nm3 (48,000 bbl) per Day Plant
Producing Naphtha and Fuel Oil from Coal
by the H-Coal Hygas Process VII-22
VII-13 Capital Investment for a 81-cm (32-in.) Diameter
Gas Transmission Pipeline—1,300 km (800 mi). . . . VII-25
VII-14 Operating Costs and Revenue Requirements
for an 81-cm (32 in.) Diameter Gas Pipeline
— 1,300 km (800 mi) VI-26
VII-15 Capital Investment for a 51-cm (20-in.) Diameter Coal
Liquids Pipeline — 1,300 km (800 mi) VII-30
VII-16 Operating Costs and Revenue Requirements for
a 51-cm (20-in.) Diameter Liquids Pipeline
— 1,300 km (800 mi) VII-31
VII-17 Capital Investment for a 270 MW Combined-Cycle
Power Plant Using Distillate Fuel from the
H-Coal Process VII-39
VII-18 Operating Costs and Revenue Requirements for a
270-MW Combined-Cycle Power Plant Using
Distillate Fuel from Coal (35% Load Factor) .... VII-40
VII-19 Fuel-Cell Stack Characteristics VII-43
VII-20 Fuel-Cell Trailer Cost Summary VII-44
VII-21 Equipment Module Cost Breakdown VII-47
VII-22 24.0-MW Fuel-Cell Power Plant Cost Estimate VII-49
VII-23 Optimistic Cost Projection for 24-MW
Fuel-Cell Power Plant VII-52
VII-24 Operating Costs and Revenue Requirements
for a 24.0-MW Fuel-Cell Power Plant VII-53
xx ix
-------
VII-25 Cost Breakdown of Naphtha Reformer Package VII-55
VII-26 Equipment Module Cost Breakdown VII-56
VII-27 25.6-MW Fuel-Cell Power Plant Cost Estimate VII-57
VII-28 Optimistic Cost Projection for 25.6-MW
Fuel-Cell Power Plant VII-59
VII-29 Operating Costs and Revenue Requirements for a
25.6-MW Fuel-Cell Power Plant VII-60
VII-30 Fuel-Cell Stack Costs VII-67
VII-31 Cost Summary for 100-kW Power Plant VII-70
VII-32 Optimistic Projection of Advanced
100-kW Power Plant VII-71
VII-33 Operating Costs and Revenue Requirements
for a 100-kW Fuel-Cell Power Plant
with Heat Recovery VII-73
VII-34 Capital Investment Required 'for a System That
Delivers 82°C (180°F) Hot Water to
Twenty Townhouses VII-75
VII-35 Capital Cost for Residential Heating and
Cooling System—Gas Furnace and Air
Conditioner VII-77
VII-36 Capital Costs for Residential Heating
and Cooling Systems — Heat Pumps VII-78
VIII-1 Monthly Light and Appliance Electrical Loads
for Three Types of Residences VIII-3
VIII-2 Normal Monthly Average Temperatures, °C (°F) .... VIII-5
VIII-3 Comparison of Normal Monthly Conditions
With Actual Monthly Conditions of Months
Chosen as "Best Match" VIII-6
VIII-4 Components of the Cooling Load — Omaha Summer
Afternoon Conditions VIII-10
VIII-5 Summary of Heating and Cooling Loads by Month VIII-11
VIII-6 Electricity Consumption for Heating and Cooling VIII-13
VIII-7 Residence 3 Electricity Consumption and
Fuel-Cell Heat Utilization VIII-23
XXX
-------
IX-1 Cost of Heating and Cooling for System 1 IX-12
IX-2 Cost of Heating and Cooling for System 2 IX-12
IX-3 Cost of Heating and Cooling for System 3 IX-13
IX-4 Cost of Heating and Cooling for System 4 IX-13
IX-5 Cost of Heating and Cooling for System 5 IX-14
1X-6 Capital Intensiveness of the System Components IX-20
IX-7 Capital Intensiveness for System 1 IX-21
IX-8 Capital Intensiveness for System 2 IX-22
IX-9 Capital Intensiveness for System 3 IX-22
IX-10 Capital Intensiveness for System 4 IX-23
IX-11 Capital Intensiveness for System 5 IX-23
IX-12 Pollutant Emissions Associated with System 1 IX-25
IX-13 Pollutant Emissions Associated with System 2 IX-26
IX-14 Pollutant Emissions Associated with System 3 IX-27
IX-15 Pollutant Emissions Associated with System 4 IX-28
IX-16 Pollutant Emissions Associated with System 5 IX-29
IX-17 Geographically Weighted Pollutant Emissions
for the Five Systems IX-32
IX-18 Occupational Exposure Standards for Toxic Pollutants
Time-Weighted Averages IX-33
IX-19 Weighting Factors for the Relative Hazards
of Air Pollutants IX-35
IX-20 Hazard-Weighted Air Pollutant Emission Factors
for the Five Systems IX-35
IX-21 Land Use Factors for System Components IX-38
IX-22 Total Land Use for the Five Systems IX-39
IX-23 Sources of Involuntary Exposure
to High Noise Levels IX-40
X-l System Rankings in Various Categories X-2
xxx i
-------
GLOSSARY OF ABBREVIATED TERMS
AN audible noise
BaP benzo(a)pyrene
COP coefficient of performance
DCF discounted cash flow
DHW domestic hot water
ESP electrostatic precipitator
FGD flue gas desulfurization
HDS hydrodesulfurization
HHV higher heating value
L&A light and appliances
NSPS new source performance standards
PAH polycyclic aromatic hydrocarbons
PFI plant facilities investment
SNG synthetic natural gas
GLOSSARY OF UNITS
bbl barrel
Btu British thermal unit
cm ...... centimeter
dB decibel
dBA decibel measured on the "A" scale
E cell potential
E theoretical cell potential
EMF electromotive force (half-cell potential
xxxiii
-------
ft foot
g gram
gal gallon
gal/min gallon per minute
GJ gigajoule
hr hour
in. inch
kg kilogram
kj kilojoule
km kilometer
kPa kilopascal
kV kilovolt
kW kilowatt
kWh kilowatt-hour
Ib pound
m meter
3
m cubic meter
mA milliampere
mi mile
MJ megajoule
mph miles per hour
MW megawatt
3
nm normal cubic meter
ppm parts per million
Psi pounds per square inch
psia pounds per square inch atmospheric
Psig pounds per square inch gauge
scf standard cubic foot
tonne metric ton
V volt
3
yd cubic yard
xxx iv
-------
I. INTRODUCTION
Within the past few years, fuel cells have received substantial
research and development attention. This R&D effort sponsored by the
Environmental Protection Agency, the Department of Energy, the Electric
Power Research Institute, the Gas Research Institute and individual
companies, this R&D effort is endeavoring to commercialize first-
generation fuel-cell power plants in the early 1980s, and to develop
advanced fuel-cell technology for applications in the late 1980s and
early 1990s.
Although fuel cells have been in existence since the early 1800s
and were used successfully in the U.S. space program as spacecraft power
plants, they have not yet become cost-competitive for terrestrial appli-
cations. Successful completion of current R&D programs would yield
clean, cost-effective, energy-efficient power plants for use in utili-
ties, industry, and buildings.
The deployment of fuel cells in dispersed electrical power plants
promises considerable operational and environmental advantage over
current methods of power generation. However, as the nation relies more
on coal and coal-based fuels for its energy sources, the effects of the
entire fuel cycle associated with new energy technologies will become
more significant. Because of its continuing interest in fuel cells as
potentially clean power sources, and its responsibility to understand
the impacts of advanced energy systems, the Environmental Protection
Agency (EPA) contracted with SRI International and its subcontractor,
Exxon Research and Engineering, to perform an assessment of fuel cells
and their associated fuel cycles.
1-1
-------
The work reported here represents the first wide-ranging assessment
of energy supply systems based on fuel cells. It addresses the environ-
mental impacts, costs, energy efficiencies, and performance character-
istics of several complete energy supply systems, including extraction
of energy resources and the end use of the delivered energy. In parti-
cular, this work analyzes the advantages and disadvantages of energy
supply systems based on fuel cells. The objective of this report is to
determine whether fuel cells and the fuel cycles that support them will
provide advantages over other systems in comparable applications. Be-
cause of the considerable potential for fuel cells to provide clean,
efficient power generation, it is important to determine whether that
potential can be realized when broader systems considerations are taken
into account relative to alternatives in similar applications.
Both fuel-cell and nonfuel-cell systems are analyzed, but only sys-
tems comparable in terms of their primary energy resource, end use, and
time frame for deployment are compared. The specific criteria used to
select the systems (as determined by EPA in the Statement of Work) are
discussed in Chapter III.
Our method of approach in the analysis was to specify the structure
of the systems in a general way, then to describe the operating charac-
teristics of the system components in detail. Next, we analyzed the
economic, efficiency, environmental, and performance characteristics of
the components. Finally, we combined the component parameters to deter-
mine the overall attributes of the systems.
The structure of this report closely follows the method of approach
and is designed to present a logical, step-by-step approach to the
description, analysis, and conclusions regarding the implementation of
energy systems based on fuel cells.
Chapter II presents an overview of fuel-cell technology, economics,
and applications. Designed to familiarize the reader with the fuel-cell
1-2
-------
concept, it provides a basis for understanding the technical discussions
of fuel cells in subsequent chapters. Electrochemical reactions, opera-
tional characteristics, and innovative fuel-cell technologies are
described.
Chapter III outlines the procedures by which five energy systems
were selected for detailed analysis. Complete energy systems included
coal extraction and transportation, coal conversion, product transpor-
tation, electricity generation and transmission, and residential energy
use. The selection procedure first limited the number of systems to 12
energy supply possibilities, which included several power plants using a
variety of coal-derived fuel types. Utilizing a careful ranking pro-
cedure, we chose five systems representative of viable, economical,
efficient energy supply systems likely to be available after 1985 but
before 2000.
Chapter IV presents detailed descriptions of the five systems,
including the particular technology chosen for each system component,
the operation of each component, and, where appropriate, flow charts,
material balances and equipment layouts. In the case of fuel-cell power
plants, detailed information is presented on design criteria, fuel-cell
performance, and power plant components.
Chapter V presents an analysis of the energy efficiency of each
system component, based on thermal balances and estimated conversion
efficiencies. Chapter VI describes the environmental aspects of the
systems components, including the specification of environmental control
equipment, rates of pollutant emissions to the air, water, and land, and
discussion of such factors as land use, aesthetics, and noise. Chapter
VII describes the economic aspects of the systems, including the capital
costs of the systems components, the cost of operating each component,
and cost sensitivity curves. In addition, detailed discussions of the
cost components of fuel-cell power plants are included.
1-3
-------
In Chapter VIII, the energy use characteristics of the residences
supplied by the systems are determined. The monthly electricity con-
sumption by lights and appliances is estimated, and a procedure for
calculating monthly heating and cooling energy use is shown.
In Chapter IX, the areas of economics, energy efficiencies, en-
vironmental impact, and overall system performance of the five systems
are analyzed and compared. Overall system performance parameters are
derived based on system information described in previous chapters.
Finally, in Chapter X, the advantages and disadvantages of each
system are summarized and conclusions are drawn regarding the relative
merits of each, particularly those that use fuel cells.
1-4
-------
II. OVERVIEW OF FUEL-CELL TECHNOLOGY
A. The Fuel-Cell Concept
A fuel cell is a device for converting the chemical energy of a
fuel directly into electrical energy, using electrochemical reactions.
Fuel-cell reactions are carried out selectively in two distinct, de-
coupled zones within a device that contains an ionically conductive
(electrolytic) medium. Such a device is shown schematically in
Figure II-l, using hydrogen as fuel, oxygen or air as oxidant, and an
aqueous acid electrolyte. The reactions (in this example) take place
within a porous electrode structure containing hydrophobic regions to
permit access of gaseous reactants and hydrophilic regions accessible to
the liquid electrolyte. Suitable catalysts may also be incorporated to
increase reaction rates.
The specific fuel-cell reactions taking place at each electrode are
as follows:
Anode reaction H2 *- 2H+ + 2e~
Cathode reaction: %0 (air) + 2H+ + 2e~ •- HO
Overall reaction: H + %0. •- HO + direct electrical energy
The result is the direct generation of electrical energy (electrons)
that can be used to produce useful work in external circuits.
Conventional energy conversion devices, on the other hand, operate
in an indirect manner, shown schematically in Figure II-2. Chemical
energy is first converted to heat in high-temperature combustion react-
ions. This heat, usually the form of steam, is converted to mechanical
work in a steam turbine and then to electrical energy using a rotating
electrical generator.
II-l
-------
02 (AIR)
Ijjiiiiiiia HEAT
ANODE
CATHODE
ELECTROLYTE
FIGURE 11-1. THE FUEL-CELL CONCEPT
FUEL
AIR
HEAT
(STEAM)
MECHANICAL
WORK
ELECTRICAL
ENERGY
FIGURE 11-2. CONVENTIONAL ENERGY GENERATION
II-2
-------
The maximum conversion efficiency of a thermal cycle is given by
the Carnot efficiency equation:
— T
Net work done ~ c
Carnot efficiency = _ . , .—-—r r—J = ™—
3 Total heat absorbed TH
where T and T are the absolute temperatures at which heat is
absorbed and rejected to the environment, respectively. The effect of
varying T on Carnot efficiency when the rejection temperature is
fixed at 30°C (86°F) is shown in Table II-l.
Typically, modern conventional steam power plants operate with a
TH of about 500 C (930 F), yielding a maximum theoretical conver-
sion efficiency of 61%. Mechanical and other losses further reduce the
efficiency to a practical limit of about 40%. Attempts to improve this
efficiency level by increasing T have been hindered by severe mater-
£1
ials problems associated with high-temperature operation. Furthermore,
production of NO pollutants increases substantially as higher com-
X
bustion temperatures are employed in these thermal systems.
Table II-l
EFFICIENCY LIMITS OF CONVENTIONAL
THERMAL ENERGY CONVERSION SYSTEMS
Heat Absorption Temperature, TJJ Carnot Efficiency*
°C (°F) (percent)
500 (930) 61
700 (1,300) 69
1,000 (1,800) 76
1,200 (2,200) 79
*Heat rejected at TC = 30°C (86°F).
II-3
-------
The direct energy generation approach used in the fuel cell is in-
trinsically more efficient. Here, an equivalent "thermal" efficiency
can be defined, using a thermodynamic characterization of the fuel-cell
reaction:
AG = AH - T AS
where AG is the reaction free energy, AH is the enthalpy of reaction
(equivalent to the higher heat of combustion), and AS is the entropy
change, with all values calculated at an absolute reaction temperature,
T. Thus, the following relationship exists for the fuel cell:
Equivalent thermal efficiency = AG/ AH = 1 - T AS/ AH
Examination of a wide range of potential fuel-oxidant electrochemical
reactions show that efficiencies approaching 100% are theoretically
possible.
This intrinsic efficiency advantage of direct energy conversion,
involving reduced fuel consumption, is the principal reason for the
ongoing interest in fuel-cell technology. In addition, fuel-cell reac-
tions are generally carried out at relatively low temperatures with
minimal NO production, in "static" configurations having few rotating
X
components. Thus, the fuel cell shows considerable promise for energy
generation with minimum adverse environmental effects such as air
pollutant emissions, and thermal or noise pollution.
B. Fuel-Cell Applications
A number of power generation applications have been proposed to
capitalize on the advantages of fuel-cell technology. The applications
that are currently under consideration for commercial development in-
clude on-site total energy systems; dispersed-site electric utility
systems; and central-site, base-load electric utility systems.
H-4
-------
Because of low emissions, fuel cells can be installed on-site very
close to a load center. In turn, this close geographic coupling of
power plant and electric demand makes recovery of the waste heat gen-
1*
erated during fuel-cell operation attractive. Such total energy or
cogeneration systems are being studied actively. Typical sites under
consideration include residential apartment complexes, commercial shop-
2 3
ping centers, and industrial facilities. '
First generation phosphoric acid fuel-cell power plants have been
proposed for this application. Typical electrical power levels range
from 40 kW to several megawatts. Waste heat recovery in the form of
steam or hot water could raise the effective utilization of fuel energy
to levels approaching 90%, as shown in Figure II-3.
100
27°C(80°F)
WATER RETURN
600C(I40°F)
WATER RETURN
0 25 50 75 100
PERCENT RATED ELECTRIC POWER
Source: Reference 2
FIGURE II-3. EFFICIENCY CHARACTERISTICS OF FUEL-CELL
POWER PLANT WITH HEAT RECOVERY
''Numbered references are listed at the end of each chapter.
II-5
-------
Fuel-cell siting flexibility also permits the installation of lar-
ger power plants at substation locations close to load centers within
electric utility grids. Such installations, rated at about 25 MW,
operating in efficient cycling or load-following modes, are expected to
have a number of beneficial effects on utility operation and economics,
including:
o Production cost credits associated with load management of
utility cyclic intermediate and peaking generation capacity.
o Efficient provision of required spinning reserve capacity;
also, many small redundant generators installed on the system
will reduce reserve requirements.
o Transmission and distribution (T&D) credits resulting from
capital cost savings due to deferred T&D expansion and reduced
line losses, as the distance between generation site and load
demand center is shortened.
Translating these benefits for installing fuel-cell power plants
yields anticipated investment credits in the range of $50-150/kW. How-
ever, such credits are very site-dependent and will vary considerably
from utility to utility. Initially, utilities serving constrained urban
markets are expected to benefit most from the use of fuel cells.
Similar benefits have also been projected for 5 to 10-MW fuel cells
installed in small municipal and rural utilities. Here again, high
efficiency under varying load output or even base-load conditions pro-
vides the incentive for introducing fuel cells.
Finally, advanced technology fuel cells, such as the molten car-
bonate or solid oxide electrolyte systems, have been proposed for
600-MW+ base-load power plants using abundant coal resources as primary
8 9
fuel. ' Coal would be gasified to form hydrogen and carbon monoxide
rich mixtures, prior to entering the fuel cell. Initial projections
indicate high energy conversion efficiency. The installed cost of such
systems, based on conceptual power plant designs, appears competitive
with alternative conversion technologies, but further study is required
to firm up these cost estimates.
II-6
-------
C. Potential Fuel-Cell Advantages and Disadvantages
The attractive characteristics of fuel cells can be classified
broadly as follows: thermodynamic, environmental, structural, opera-
tional, and economic. Some characteristics are intrinsic to the concept
of "fuel cell;" others are derivative, based on practical operating de-
vices. Naturally, considerable overlap occurs and some attributes show
up in several categories.
The major reason for initial and continued interest in fuel cells
is their intrinsic thermodynamic advantage. The high efficiency of fuel
cells, which conserves natural fuel resources, is not a functon of fuel-
cell plant size or power output level, and can be obtained in a variety
of potential applications, from small dispersed apartment complex units
to large central utility units (see Figure II-4, which compares the ef-
ficiency of fuel cells with other competing technologies).
High efficiency can be maintained at rated or part load, so that
fuel cells have attractive load-following characteristics (see Figure
II-5). Furthermore, the efficiency can be enhanced-by the use of the
heat generated during fuel-cell operation, particularly in total energy
concepts with a thermal load demand in addition to the usual electrical
load demand.
Fuel cells are projected to be low-pollution energy conversion
devices. Low levels of air pollution are attained through the selecti-
vity and completeness of the clean fuel oxidation reaction, compared
with the uncontrollable high temperature combustion reactions involved
in conventional devices. Thus, low temperature fuel-cell reactions do
not produce the CO and NO by-products characteristic of normal com-
X
bustion. The waste products of fuel-cell operation are generally harm-
less and nonpolluting water and carbon dioxide.
High efficiency operation results in less waste heat, so that the
thermal pollution burden is low. Also, many fuel-cell systems reject
II-7
-------
Si
ui 5
5»
u. £
fc|
£1
|o
111 **
0 I
STEAM AND
GAS TURBINE
SYSTEMS
RATED POWER OUTPUT - kW
Source: Reference 4
FIGURE 11-4. EFFICIENCY OF VARIOUS ENERGY CONVERSION DEVICES
I
IU
o
Z
IT
Ul
DC
20 30 40 50 60 70 80
PERCENT OF RATED POWER OUTPUT
Source: Reference 4
90 100
FIGURE 11-5. PART-LOAD EFFICIENCY
II-8
-------
waste heat to the atmosphere, not to surface water systems. Thus,
thermal pollution to adjacent water bodies is eliminated.
In addition to having low pollutant releases, fuel-cell design and
assembly is flexible, and unobtrusive power plants can be designed hav-
ing wide siting capability that makes effective use of available land
while maintaining an aesthetically acceptable appearance.
The structural advantages of fuel cells result from modular,
filter-press construction, which is the current technology approach to
solving the complex reactant segregation and electrochemical constraints
of most fuel cells. Modular building blocks allow the same technology
to be packaged for widely different market needs, over a wide range of
voltage and power levels, watts to kW to MW, for portable to transport-
able to stationary applications. Small modular units imply high unit
sales volume, which can result in high production rates, low-cost fac-
tory assembly, and improved quality control. Moreover, modular assembly
will bring about reduced lead time for ordering parts and installing
total generating plants, reduced installation and check-out times, and
increased ease of installation. Therefore, field construction costs can
be minimized.
Rapid installation of basic modules in prefabricated increments is
responsive to demand growth. Self-contained modules can be placed in
rapidly gowing demand areas, including existing communities with under-
ground constraints or site inflexibility. Modular elements can be as-
sembled and used in various forms and spatial geometries, according to
application needs and volume constraints. Compact, lightweight power
plants are possible, depending on reaction kinetic factors and reactants
used, leading to high power density and high energy density.
The expected operational advantages of fuel cells can be divided
into those applying to the fuel cells themselves, and those applying to
the technology that will be based on them. Fuel cells generally contain
multiple units that are easily operated in parallel, and are built in
II-9
-------
modular units, allowing power availability even though a portion of the
plant is off-line; remaining modules can temporarily increase output
without major efficiency or operating cost penalty. A modular subsystem
gives redundancy and thus increases overall reliability.
Fuel cells are expected to provide ease of maintenance with re-
sulting low cost. Balanced rotation and unstressed moving parts in
auxiliary equipment give low maintenance, good reliability, and long
life. They operate at a constant temperature, so that lower thermal
stresses are exerted. They also contain simple elements that are easily
operated or stored, and that offer automatic operation capability, ease
of startup, and remote site operation capability.
Fuel-cell technology in its various applications can offer dis-
tributed, efficient power generation close to load centers, provide
power generation capability involving minimum interactions with other
utility functions, and provide transportable power generation in such
emergencies as flood, earthquake, and hurricane. It is also adaptable
to energy storage schemes coupled with periodic generation (e.g., solar)
via a hydrogen electrolysis/conversion operation.
The future trend in alternative power generation technologies is
toward higher cost due to rising fuel prices, labor rates, and cost of
debt capital, and the imposition of new, higher cost, environmental
regulations. This trend will improve the competitive stance of fuel
cells because fuel-cell technology is still developing and prospects
exist for improved life-cycle costs with the potential for reasonable
investment costs, lower installation, operation, and maintenance costs
over a period of years.
A particular economic advantage of fuel cells is that they could
add utility service capability incrementally, in phase with actual
demand. This feature should minimize financial risks and costs, and
provide cash flow advantages by eliminating the delay time during which
demand must increase to meet increased installed capacity. Fuel cells
11-10
-------
also have a short lead time and high availability of plants versus
existing alternatives. In addition, their flexibility permits instal-
lation closer to consumer, reducing transmission/distribution costs and
utility service distribution infrastructure costs.
At present, the principal disadvantage of fuel-cell technology is
its relative newness. Few, if any, complete power plants have been
constructed, installed, and tested for prospective users to assess the
expected operating advantages of fuel cells. This situaton will change
as ongoing programs provide demonstration systems. However, consider-
able uncertainty still exists concerning the eventual installed cost,
durability, and lifetime of mature fuel-cell technologies. Fuel cells
will have to compete with other, more established, forms of energy con-
version. These alternative systems are also being improved so that
fuel-cell technology has to match moving performance and cost targets.
Earlier fears about the availability of liquid hydrocarbon supplies
used by first-generation (phosphoric acid) fuel-cell systems have eased
recently. In any event, the high conversion efficiency of fuel
cells should represent the most effective use of potentially scarce fuel
resources.
D. The Complete Fuel-Cell System
The complete fuel-cell system consists of an integrated modular
assembly of subsystems including the fuel conditioner for converting
primary fuels to hydrogen; the fuel-cell stacks containing planar
electrodes and an electrolyte-holding matrix as sandwiched sheets,
housed in plate-and-frame filter-press assemblies; and a power condi-
tioner for converting the primary fuel-cell electric output (direct
current) to alternating current at a suitable voltage. The inter-
relationship of these components is shown in Figure II-6.
11-11
-------
FUEL-
Air
FUEL
CONDITIONER
L
FUEL CELL
DC
POWER
CONDITIONER
AC
Spent Fuel
FIGURE 11-6. THE FUEL-CELL SYSTEM
These modular assemblies can be scaled in size for any power level
(kW to MW) application. In general, a high level of factory assembly of
individual components into shippable arrays is projected, which will
facilitate power plant installation in the field. The key factors in
the design and operation of a complete fuel-cell power plant are fuels
(or reactants) to be used, the fuel-cell operating characteristics, the
overall economics of power plant operation, and the technology trade-
offs among types of fuel-cell stacks, operating conditions, lifetime,
and so on. These factors are discussed in the following sections.
1. Fuel-Cell Reactants
A wide range of reactants, particularly fuels, can be considered
for use in fuel-cell power plants. In principle, many fuels will
undergo direct electrochemical oxidation at the fuel-cell anode.
Possible fuels include hydrogen, methanol, and hydrocarbons such as
methane. The anode (fuel) reactions are:
CH3OH
CH,
H2
H20
2H20
2H
6H
8H
2e
C0
6e~
8e~
11-12
-------
In general, only oxygen (from ambient air) is considered a
practical oxidant. The cathode (air) reaction is:
2H+ + 2e
A characteristic half-cell potential (EMF) is associated with each of
these reactions, as shown in Figure II-7. The difference between the
anodic (fuel) and the cathodic (()„) half-cell potentials is the
theoretical cell potential, E°. For example, E° for the H2/02
fuel-cell couple is +1.23 - 0.0 = +1.23V (at 25°C). E° may also be
computed from the overall change in free energy of reaction, G:
E° = - AG/nF
where n is the number of electrons involved in the reaction and F is the
Faraday constant, all expressed in consistent units.
However, the kinetics of electrochemical reactions must also be
considered in selecting appropriate fuels for fuel cells. Once again, a
number of trade-offs are involved, including:
o Fuel availability or cost
o Electrochemical activity
o Fuel storage and logistic factors.
As indicated in Table II-2, the least expensive, most easily stored
fuels are not very reactive. Hydrogen, on the other hand, is very
reactive, but costly to store and transport.
This fuel convenience, cost, and reactivity mismatch has been
resolved by incorporating a fuel conditioning device in the fuel-cell
system that converts unreactive fuels into reactive hydrogen. In the
steam reforming reactor, the most widely used of such devices, primary
hydrocarbon fuels are reacted with water (steam) to produce hydrogen for
the fuel cell via the following endothermic steam-reforming reaction:
CH4 + 2H20 Catalvst, 4H2 + C02
Heat
11-13
-------
+ 1.3 --
o
o
in
CM
<
O
HI
I-
o
a.
>
I-
lil
DC
+ 1.2 - -
//
+0.2 - -
+0.1 - -
0.0 - - -*
-0.1 - -
-0.2 - -
02 + 2H+ + 2e" = H20
C
CH4
C6H14
CH3OH
Ho 2hT + 2e"
HCHO
HCOOH
VOLTS
FIGURE II-7. THE EMF SERIES
11-14
-------
Table II-2
FUEL SELECTION TRADE-OFF
Relative Relative
Fuel Delivered Cost Electrochemical Activity
Hydrogen High Very high
Methane Low Low
Methanol Moderate Moderate
Ethanol Moderate Low
Hydrocarbons Low Low
Coal Very Low Very Low
The reaction is carried out at an elevated temperature (700°C or
1,290 F) in a fired tubular reactor packed with nickel-based reforming
catalyst. Heat for the reaction is generally provided by burning spent
anode fuel gas containing unreacted hydrogen. Because nickel-based
catalysts are susceptible to sulfur poisoning, sulfur levels in the feed
to the reactor must be kept below 1 ppm. If organic sulfur compounds
are present in the primary feed, they must be converted to H-S in a
hydrodesulfurizer using, say, colbalt-molybdenum catalyst. The H-S is
then adsorbed in a replaceable zinc oxide guard bed. Also, excess steam
is added to the reformer feed to limit the amount of by-product carbon
deposited on the catalyst bed. Typical steam/carbon molar ratios are 3
and 4 to 1.
Lastly, for fuel-cell systems using noble metal (e.g., platinum)
anode catalysts, the level of carbon monoxide in the reformer effluent
must be kept low (below 1%) to prevent excessive performance loss caused
by carbon monoxide adsorption. Carbon monoxide content can be con-
trolled using additional catalytic reactors to establish a favorable
water-gas shift reaction equilibrium:
This sequence of conversion steps can be used with hydrocarbon
feeds through the naphtha boiling point range. Heavier liquid feed
stocks have excessive carbon- forming tendencies. For these fuels,
11-15
-------
alternative conversion reactions such as partial oxidation can be used
to produce hydrogen. However, the use of air as a reactant results in
excessive nitrogen diluent in the hydrogen product gas. Similarly, coal
can be gasified using oxygen or air and steam to form hydrogen-rich
mixtures suitable for use in certain fuel-cell systems.
The interaction between the fuel and fuel cell is further compli-
cated by the availability of intermediate fuels that could be produced
from primary fuel resources, such as coal. For example, storable metha-
nol fuel could be produced from coal for certain applications where the
logistical advantages of liquid fuels are required. Further discussion
of fuel interchangeability is beyond the scope of this brief overview,
however.
2. Fuel-Cell Operating Characteristics
Although the theoretical or open-circuit potentials of a fuel cell,
E , are high, certain irreversible voltage losses are encountered when
practical levels of current are produced during fuel-cell operation.
These polarization or overvoltage losses can be characterized as follows:
Activation Polarization. Electrochemical reactions proceed at
finite rates in a number of elementary steps such as activated
adsorption/desorption and electron transfer. A potential
loss, termed activation polarization, is associated with the
slow step in the reaction sequence. This loss is small with
reactive fuels such as hydrogen, but can be quite large for
the oxygen reduction (consumption) reaction, particularly for
low-temperature fuel cells. Activation polarization losses
can be reduced by using appropriate electrocatalysts. In
general, these losses are small at the temperatures (600°C+
or 1,100°F+) employed in high-temperature fuel-cell systems.
Concentration Polarization. Fuel cells generally use porous
electrode structures to maximize the surface area available
for electrochemical reaction. Furthermore, maximum
utilization of reactants is sought. As a result, reactant
concentration gradients are established throughout the
electrode structure and flow-path through the cell. These
gradients, and the resulting reactant and product diffusion or
11-16
-------
mass transfer requirements, cause a polarization loss
associated with the effect of local concentrations on
potential. This concentration polarization loss can be
estimated using the Nernst equation:
E = E° - (RT/nF)ln [(Products)/(Reactants)]
where E is the cell voltage corrected for concentration
differences between standard state and actual cell conditions,
R is the gas constant, and the terms in the logarithm are
suitable concentration (activity) terms. The Nernst equation
can also be used to estimate the effect of high-pressure
fuel-cell operation. Such operation is usually beneficial.
Concentration polarization losses occur in all fuel-cell
systems, including the high-temperature molten carbonate and
solid oxide electrolyte systems.
o Ohmic Polarization. The movement of ions and electrons
requires a potential gradient. In turn, an ohmic polarization
loss is incurred that is directly proportional to the current
flow (ionic or electronic) and pathway resistance. Proper
selection of electrolyte composition and design and assembly
of fuel-cell structural components can minimize, but not
eliminate, ohmic polarization losses.
The impact of these polarization losses is to reduce the fuel-cell
terminal voltage. Because these losses increase as load current is
increased (see Figure II-8), fuel cells exhibit a characteristic
performance curve (voltage vs. current load) shown in Figure II-9. In
turn, the power (P=EI) delivered by a fuel cell varies with current load
(see Figure 11-10). Although high power operation is desirable, such
operation results in reduced energy conversion efficiency. A modified
efficiency parameter, the voltage efficiency, can be used to
characterize this effect:
TT i *£• • Actual cell voltage _ .__
Voltage efficiency = -:— — f- = E/E°
& J Theoretical cell voltage
Voltage efficiency decreases somewhat as current load is increased
(see Figure 11-11). For a given power plant, cell and system efficiency
increase as load output is reduced.
11-17
-------
— E°
UJ
O
O
O
UJ
O
CELL TERMINAL
VOLTAGE
ANODE
LOSSES
RESISTANCE
LOSS
CURRENT - I
FIGURE 11-8. FUEL-CELL LOSSES
CURRENT LOAD - I
FIGURE 11-9. CHARACTERISTIC FUEL-CELL PERFORMANCE CURVE
11-18
-------
X
UJ
I
DC
UJ
CURRENT - I
FIGURE 11-10. FUEL-CELL POWER OUTPUT
o
UJ
UJ
I
o
UJ
o
111
o
3
o
CURRENT - I
FIGURE 11-11. EFFECT OF CURRENT LOAD ON VOLTAGE EFFICIENCY
11-19
-------
3. Fuel-Cell Economics
The cost of energy delivered by a power plant is the sum of three
factors: capital charges, fuel charges, and operating and maintenance
(O&M) charges. Estimations of the cost of energy and selection of
strategies to reduce it are ongoing concerns of fuel-cell R&D programs.
Associated with each power plant is a total installed cost that must be
invested by the user. This investment cost must be amortized over the
operating life of the power plant. Capital charges for this investment
can be estimated from:
, , , Investment cost ($/kW) ,. 1/>rv
Capital charges (C/k»h) - Annual operating hours x fCR x 10°
where f is a capital recovery factor that includes depreciation,
C*K
taxes, overhead, profit, and other costs associated with owning the
power plant. A typical value for electric utility operation is
f,,., = 0.18. Clearly, as shown in Figure 11-12, capital charges can be
L»K
minimized by reducing the installed cost and by increasing the annual
operating time (load factor).
Fuel charges include the cost of fuel consumed by the fuel cell to
deliver a specified amount of electrical energy. It is directly related
to the delivered cost of fuel based on the higher heating value (HHV) of
the fuel. Fuel charges also are a function of the power plant
conversion efficiency. Here, system overall efficiency is expressed as
the heat rate (i.e., the amount of fuel consumed per kWh of electrical
energy delivered to the busbar). In addition to the efficiency of the
fuel cell itself, some losses are associated with standby fuel
requirements and power plant auxiliaries such as pumps, fans, blowers,
and power conditioning equipment. A power plant operating at 100%
efficiency would have a heat rate of 3,600 kJ/kWh (3,413 Btu/kWh).
Thus, the fuel charge is the product of fuel cost and heat rate. This
relationship is shown in Figure 11-13 for typical values of fuel cost.
11-20
-------
10
w
LU
O
DC
<
O
Q_
<
O
CR
0.18
POWER PLANT
INSTALLED COST
($/kW)
800
I
2000
4000 6000
ANNUAL OPERATING TIME - hrs
8000
FIGURE 11-12. CAPITAL CHARGES
7000
HEAT RATE - Btu/kWh
8000 9000
10,000
11,000
1
CO
LU
13
OL
<
O 2
FUEL COST
$/GJ ($/106 Btu)
4 (4.22)
3 (3.17)
2 (2.11)
1 (1.06)
I
T
I
I
6000
7000
8000 9000
HEAT RATE - kJ/kWh
10,000
11,000
FIGURE 11-13. FUEL CHARGES
11-21
-------
O&M charges are associated with the routine operating and main-
tenance functions normally required to run a reliable power plant. They
are usually directly proportional to the operating time at a given power
output. In addition, fuel-cell power plants are expected to incur cost
related to the periodic replacement of spent fuel-cell stacks, which can
conveniently be treated as an operating expense directly proportional to
the net stack replacement cost and inversely proportional to the stack
life. In this study, net stack replacement costs include the stack cost
itself, together with required installation labor costs, minus the sal-
vage value of the replaced stacks. Experience with spent stack recovery
is insufficient to project this salvage value, but most of the noble
metal catalyst content of low-temperature fuel-cell stacks is expected
to be effectively recovered.
The effect of stack life and net replacement cost on the O&M
charges is shown in Figure 11-14. Clearly, stack durability or life has
a negligible impact on costs for lifetimes exceeding 40,000 hours.
Accordingly, this value is usually selected as a goal for R&D
demonstration programs.
Selection of an optimum design point for the fuel-cell power plant
involves analysis of each cost factor discussed above. The effect of
load current on the cost of energy is shown schematically in Figure
11-15. Capital charges decrease as current (therefore, power output)
increases. At the same time, fuel charges will increase as cell voltage
efficiency decreases at high load. O&M charges should be relatively
constant, assuming that stack life is independent of design current
load. Generally, the total cost of energy shows only a moderate effect
of operating load.
Determining the expected installed cost of fuel-cell power plants,
together with demonstration of target stack durability, is currently
very difficult because very few detailed descriptions of fuel-cell
system capital costs have been published. Expected or cost goals for
11-22
-------
2.5
2.0
1.5
I
LU
DC
O
1-
LU
O 1.0
5
O.
LU
DC
U
0.5
NET STACK
REPLACEMENT
COST |$/kW)
200
150
100
50
I
10
20 30
STACK LIFE - 1000 hr
40
FIGURE 11-14. STACK REPLACEMENT CHARGES
50
EC
LU
LU
O
I-
o
O
TOTAL
FUEL
CAPITAL
O&M
CURRENT LOAD - I
FIGURE 11-15. DESIGN POINT SELECTION
11-23
-------
mature technology are listed in Table II-3. Recent estimates of the
cost of dispersed site phosphoric acid fuel-cell systems, at initial
production levels, range between $500-700/kW.
In summary, the greatest hurdle facing fuel-cell technology is the
simultaneous achievement of interrelated goals for performance (power
output and efficiency), durability (life and reliability), and cost.
Table II-3
ESTIMATED INSTALLED INVESTMENT COSTS
FOR FUEL-CELL POWER PLANTS
Fuel-Cell System
First-generation phosphoric
acid (40 kW+)
First-genertion phosphoric
acid (5-26 MW+)
Second-generation molten
carbonate (5-26 MW+)
Second-generation molten
carbonate with coal
gasifier (600 MW+)
Application
On-site total
energy
Dispersed site
Utility
Dispersed site
Utility
Central site
Utility
Estimated or Target
Installed Cost, $/kW
(1980$)*
400-5001
35011
28011
8348
Escalated at 7%/year from original estimate year,
4. Fuel-Cell Technologies and Trade-offs
Many technological approaches have been studied to reduce the
fuel-cell concept to a practical device. These approaches can be
classified according to the electrolyte composition and a corresponding
operating temperature level, as shown in Table II-4. Definition of an
optimal choice is difficult because the following practical trade-off
factors must be considered:
11-24
-------
Reactant Activity. In general, electrochemical reactions
proceed more rapidly at elevated temperature.. Thus, higher
power outputs are possible at high temperatures, reducing the
need for catalysts to promote the reactions. In fact,
high-temperature (600°C+, 1,100°F+) fuel cells do not
require the expensive platinum catalysts required by
low-temperature technologies.
Materials Life and Cost. On the other hand, high-temperature
operation also increases the rate of corrosion reactions,
imposing problems of materials stability and cost. For
maximum cost-effectiveness, fuel-cell power plants have to
function for several years. Thus, the question of component
durability is important.
System Application. Fuel-cell power plants have been proposed
for many divergent power generation applications, including
small-scale total energy systems; large-scale, intermittent
power dispersed-site systems; and even larger, central-site
base-load systems, as well as specialty aerospace and
terrestrial applications. Each application poses unique
requirements for which certain fuel-cell technologies may be
more suitable than others.
Electrolyte Properties. Fuel-cell electrolytes must have
reasonably high ionic conductivity. This requirement is met
by the composition-temperature combinations listed in Table
II-4. In addition, the electrolytes must be stable and
invariant with the reactants and trace contaminants
encountered in fuel-cell operation.
Table II-4
FUEL-CELL TECHNOLOGY CLASSIFICATIONS
Electrolyte
Type
Solid Polymer
Aqueous
Molten
Solid
Example
Ion exchange membrane
Acid: H3P04*
Base: KOH
Carbonates:
Oxides:
Zr02/Y203
Operating Temperature
Range, °C (°F)
70-100 (160-210)
170-200 (340-390)
70-120 (160-250)
600-700 (1,100-1,300)
1,000-2,000 (1,800-3,600)
^Technologies undergoing major development effort at this time.
11-25
-------
E. Specific Fuel-Cell Technologies
A number of fuel-cell technologies are available for development
into power plant systems. This section describes the status of specific
fuel-cell technologies currently under study in major development
programs, including:
o First-generation technology: phosphoric acid fuel cells
o Second-generation technology: molten carbonate fuel cells
o Third-generation technology: solid oxide fuel cells.
The status of other technologies such as alkaline and ion-exchange
membrane fuel cells is also reviewed briefly.
1. Phosphoric Acid Fuel Cells
Phosphoric acid fuel cells are under intensive development by
United Technologies Corporation (UTC), Energy Research Corporation/
Westinghouse, and Engelhard Industries for both on-site or dispersed-
site power generation applications. This technology uses a concentrated
phosphoric acid electrolyte, operating at temperatures up to 190°C
(374°F) at pressures ranging from atmospheric to about 345 kPa (50
psia). The electrolyte is contained within a silicon carbide matrix
sandwiched between the electrodes. Platinum electrocatalysts are
employed, in the form of highly dispersed crystallites supported on a
carbon substrate. Total cell catalyst loading is less than 1 mg Pt/cm2
of electrode geometric area. Cell cooling is accomplished by water
evaporation in isolated tubes or by flowing excess air through some of
the cathode compartments. Graphite is the principal material of
construction.
Natural gas and liquid hydrocarbons through the naphtha boiling
range are the usual primary feeds to the fuel-conditioning section of
the power plant. These fuels are converted to a hydrogen-rich feed gas
11-26
-------
using the steam reforming reaction. In some cases, by-product hydrogen
from other sources has also been considered.
A number of trade-offs are involved in selecting the operating
conditions of the phosphoric acid fuel cell. Performance (voltage and
current, hence power) improve as temperature and pressure are in-
creased. UTC is exploring the effects of increased temperature (200°C
or 392°F) and pressure (640 kPa or 93 psia). The impact of trace
carbon monoxide content in the reformed gas feed is reduced at high
temperature. Operation at high pressure, hence reduced gas flow
volumes, reduces the extent of phosphoric acid vaporization that occurs
at high temperature. However, efficient system operation at elevated
pressure requires the use of a close-coupled compressor/expander set for
the air stream. Cost-effective devices are available for the MW-sized
fuel-cell systems, but not for the smaller on-site power plant systems.
Excessive electrolyte loss limits fuel-cell durability (life), but UTC
has recently developed electrode structures that contain an electrolytic
reservoir to compensate for vaporization losses.
Air electrode (cathode) operation appears to be the major source of
concern regarding phosphoric acid fuel-cell life. The following cathode
performance degradation mechanisms have been identified and are
currently under intensive study:
Platinum catalyst sintering or surface area loss resulting in
reduced catalytic sites for the oxygen reduction reaction.12
Recent UTC studies suggest that the larger platinum
crystallites may be more accessible to reactants, compensating
for the loss in the catalyst area.
Platinum dissolution at high cathode potentials, such as
open-circuit or low-current load operation.13 xhe dissolved
platinum species migrate towards the anode (H2 electrode),
thus reducing the number of active sites in the cathode. In
effect, this places a limit on the maximum cell voltage and
efficiency achievable by phosphoric acid fuel-cell
technology. The heat rate goal for this technology is
9,800 kJ/kWh (9,300 Btu/kWh) at end of life, corresponding to
a cell terminal voltage somewhat below 0.7 volts.
11-27
-------
o "Corrosion" or oxidation of the carbon support on which
platinum catalyst is dispersed^, which results in subtle
changes in electrode structure, thickening the electrolyte
film through which dissolved oxygen must diffuse to reach a
catalytically active site. A search is under way for more
stable carbon supports.
There are no reports in the literature indicating that target life
(40,000 hours) at design performance levels has been achieved in
phosphoric acid fuel-cell stacks. Demonstration of this goal will be a
key milestone.
The use of platinum electrocatalysts in this technology appears
cost-effective. Suitable procedures for recovering catalyst from spent
stacks must be developed. The search for alternative catalyst composi-
tion, less susceptible to sintering, poisoning, or dissolution has been
attempted in the past without notable success. The extremely corrosive
nature of the electrolyte and cell-operating conditions severely limits
the range of potential substitutes for platinum.
Studies are also under way to develop acidic electrolyte substi-
tutes for phosphoric acid. Fluoro-sulfonic acid derivatives appear to
give improved cathode performance at lower temperature, but problems
with stability and/or excessive vapor pressure have been encountered.
Phosphoric acid fuel cells are prime candidates for near-term
commercialization. A number of technology demonstration programs are
nearing the critical phase. These include construction, installation,
and testing of several 40-kW power plants for on-site total energy
systems throughout the United States and a 4.8-MW demonstrator system to
be installed in New York City on the Consolidated Edison grid. If these
demonstration programs are completed successfully, the anticipated
benefits of fuel-cell technology will be confirmed.
11-28
-------
2. Molten Carbonate Fuel Cells
The molten carbonate cell has been selected for development as the
second-generation fuel-cell system. Substantially improved electro-
chemical performance is projected for this high-temperature cell
(650°C or 1,200°F). Target heat rate for hydrocarbon-fueled
intermediate-load power plants is about 7,900 kJ/kWh (7,500 Btu/kWh)
compared with 9,800 kJ/kWh (9,300 Btu/kWh) for first-generation
phosphoric acid fuel-cell systems. In addition, recent studies by
o
UTC showed that a base-load, integrated coal gasifier/molten
carbonate fuel-cell system might operate at energy-conversion
efficiencies approaching 50% (7,400 kJ/kWh or 7,000 Btu/kWh).
Significant progress has been made in recent years to overcome the
problems encountered in earlier molten carbonate fuel-cell studies. The
molten carbonate fuel cell is under active development by UTC, Energy
Research Corporation, the Institute of Gas Technology, General Electric,
and several National Laboratories.
The molten carbonate fuel cell contains a lithium aluminate matrix
or tile impregnated with a mixed potassium/lithium carbonate electro-
lyte, sandwiched between two porous sintered-nickel electrodes. The
carbonate ion participates actively in the electrode reactions:
Anode reaction: H2 + COJ "- H20 + C02 + 2e~
Cathode reaction: JgO + CO +2e~ •- COJ
Overall reaction: H_ + %02 *- H^O + direct electrical energy
Typical fuel gases include steam-reformed hydrocarbons (containing
mostly H with some CO.) and gasified coal (containing roughly
equimolal mixtures of HZ and CO, with N, diluent if air is used as
gasifier feed). At the elevated cell temperatures, about 650°C
(1,200°F), hydrogen electro-oxidation proceeds rapidly on the nickel
electrode surface. Direct electrochemical reaction of CO probably plays
a minor role, but substantial quantities of additional H_ are
11-29
-------
generated via the water gas shift reaction that also takes place on the
nickel electrode:
On the other hand, kinetics of the oyxgen reduction reaction on the
nickel (oxide) electrode are not very rapid, so that some activation
polarization loss is encountered. Ionic-resistive losses in the
molten carbonate electrolyte are another major source of performance
loss. These losses can be reduced by operating with thinner electrolyte
tiles or by increasing the temperature. However, the former approach
increases the risk of reactant gas leakage or crossover and the latter
aggravates the materials problems encountered with corrosive carbonate
melts.
Large-scale, coal-fueled molten carbonate fuel-cell power plants
are projected to operate at elevated pressure, 1,034 kPa (150 psia),
with 85% utilization of fuel (H_ + CO) and 50% oxygen utilization per
pass through the cell. To preserve electrolyte invariance, carbon
dioxide must be supplied to the air electrode. At present, burning the
spent anode gas conveniently accomplishes this through use of cathode
air fuel. The response of molten carbonate fuel-cell performance to
variations in composition can be estimated from the Nernst equation:
RF .
E = ' n? ln
(
a
1_1
I c c J
where the terms in parentheses are reactant partial pressures and the
subscripts refer to anode (a) and cathode (c) conditions. Increased
pressure operation improves cathode performance, but slightly decreases
anode performance. Recent evidence suggests there may be a beneficial
effect of pressure on reaction kinetics as well. However, high-pressure
operation may also lead to formation of inert methane, resulting in a
loss of fuel utilization:
3H + CO
or: *
11-30
-------
Molten carbonate fuel cells are usually cooled by recirculating the
cathode air stream to a separate heat exchanger. Waste heat from the
cell is thus available at a very attractive temperature (600°C or
1,100°F) for further use. For example, heat can be recovered as steam
for use in cogeneration facilities or for use in a steam-bottoming cycle
for additional power production. In this manner, substantial quantities
of energy can be generated, greatly improving the overall fuel-to-
electricity conversion efficiency. In a UTC design for a 635-MW
integrated coal-fueled power plant, the steam-bottoming cycle provided
one-third of the delivered power output."
The performance of molten carbonate fuel cells has improved
impressively in recent years, as shown in Figure 11-16. Nevertheless,
considerable progress is still required to achieve acceptable durability
or life in multicell stack operation. Individual cells have operated
stably, at reduced performance levels, in excess of 12,000 hours. The
corresponding value for a 19-cell stack of 0.093 m^ (1 ft^) cells is
about 1,500 hours.15 The following life-limiting phenomena were
recently reviewed in an EPRI-sponsored workshop:18
o Electrolyte Tile Stability. The crystal structure of the
lithium aluminate matrix changes with time. Density
differences among the different crystal types can produce
reactant gas crossover paths, particularly if the tile is
thermally cycled. Such cycling is expected in typical molten
carbonate fuel-cell system applications. The result is
reduced reactant utilization and lower cell performance. In
extreme cases, explosive gas mixtures may be formed. This
problem might be aggravated by projected operation at elevated
pressure and by the desire to reduce tile thickness to achieve
lower internal ohmic losses.
o Wet Seal Corrosion. At present, the edges of the cell are
sealed by pressing the nickel electrode against the tile
containing the mixed carbonate electrolyte. The resulting
seal develops a surface with air on one side and fuel gas on
the other, leading to local corrosion reactions. Alloys and
surface coatings containing chromium or aluminum are under
study to solve this problem.
o Electrode Sintering. Long-term operation of earlier porous
nickel electrode structures showed that loss of surface area
occurred via a sintering mechanism. This problem appears to
have been solved by the addition of stabilizing components to
the nickel.
o Materials Stability. Recent cost estimates for large-scale
molten carbonate fuel-cell systems have assumed that stainless
steels can be used as current collectors and spacers in the
cell. Long-term stability of these steels has not yet been
demonstrated. Nickel can be used as an alternative
construction material at somewhat greater cost.
11-31
-------
w
M
LL <
mE
'
z <
o ui
en ^
UJ —
160
120
80
40
1977
i
102
103
104
105
OPERATING TIME - hr
Source: Reference 17
FIGURE 11-16. PROGRESS IN MOLTEN CARBONATE
FUEL CELL PERFORMANCE
11-32
-------
o The Carbon Problem. Initial designs for the integrated coal
gasifier/molten carbonate fuel-cell power plant resulted in a
projected fuel gas feed that is thermodynamically unstable
with respect to carbon deposition:
2 CO———C02 + C |
Subsequent cell tests have verified that the nickel electrode
surface is an effective catalyst for this reaction, resulting
in carbon deposition and cell blockage. Proposed solutions to
this problem involve reducing the CO partial pressure at the
cell inlet by means of steam injection or anode gas recycle.
Fortunately, the feed gases produced by hydrocarbon steam
reforming are thermodynamically stable.
o The Sulfur Problem. Recent experiments have confirmed a
suspected deleterious effect of sulfur on molten carbonate
fuel-cell operation. Sulfur compounds (l^S, COS,
mercaptans) are present in feed gases derived from high-sulfur
primary fuels such as coal or heavy liquid hydrocarbons.
These sulfur compounds can react with the nickel anode to form
nickel sulfides, some of which are molten at cell-operating
temperatures. The result is a substantial loss in
performance. In addition, if spent fuel is burned with
cathode air feed to provide the required C0£ recycle, the
resulting S02 will be absorbed by the carbonate
electrolyte. Again, performance losses are expected.
At present, it appears that the maximum inlet sulfur content
will be limited to a few ppm. These low concentration levels
can be achieved by existing gas purification processes, but at
extra cost and, perhaps, loss of efficiency. On the other
hand, coupling the molten carbonate fuel cell with current
steam-reforming processes should not be a problem because
extensive feed desulfurization is already required to protect
the nickel-based steam-reforming catalyst.
This discussion of molten carbonate fuel-cell technology has
focused on current mainline programs. Another approach is being
studied. SRI International is conducting an exploratory effort to
define the feasibility of converting carbon (coal) directly to
19
electrical energy in a form of molten carbonate fuel cell. Here
carbon anodes are consumed electrochemically. Preliminary cost studies
are under way to define the economic feasibility of this approach.
The molten carbonate fuel cell holds considerable promise as an
advanced energy conversion system. Improved energy conversion
11-33
-------
efficiencies are obtained at the higher cell temperatures. Coupling the
molten carbonate fuel cell with primary fuel provides a good match for
the CO £ balance required for the fuel-cell operation. Thus, the mol-
ten carbonate fuel cell could become dominant in large-scale, coal-
fueled, base-load power generation. Applications involving substantial
thermal cycling may not be attractive, if current problems with tile
stability remain unresolved.
3. Solid Oxide Fuel Cells^
A further improvement in the efficiency of large-scale fuel cells
is projected through the use of solid oxide electrolyte fuel cells.
Here, advantage is taken of the ionic conductivity shown by certain
oxides, such as yttria-stabilized zirconia, at very high temperatures —
about 1,000 C (1,800 F). Reaction rates at these temperatures are
rapid, so that ohmic and concentration polarization are the principal
sources of performance loss. Apparently, carbon monoxide can also react
electrochemically at 1,000°C, an advantage when the solid oxide cell
is fed with gasified coal (H + CO).
Solid oxide cell electrochemical reactions can be written as:
Anode reactions: H2 + 0= •- H20 + 2e~
and/or : CO + 0= •- 0)2 + 2e~
Cathode reaction: ^0 + 2e~ »- 0=
2
Overall reactions: H2 + %02 — H20 + direct
electrical energy
and/or : CO + h02 •- CO 2 + direct
electrical energy
Oxide ion is the ionic current carrier, presumably moving through
vacancies in the anionic lattice of the mixed zirconia oxide crystals.
As written, the electrolyte appears invariant.
11-34
-------
Structurally, the solid oxide cell differs considerably from other
fuel-cell technologies. To form a total cell, thin layers of active
material are deposited on the outer surface of a porous ceramic support
tube. Fuel gas is fed to the tube interior and air to the exterior.
Currently, active cell components include a nickel-zirconia cermet
anode, yttria-stabilized zirconia electrolyte, a doped indium oxide
cathode, and a lanthanum-chromite interconnect containing aluminum ions
for improved thermal expansion characteristics and magnesium ions for
improved electronic conductivity. The mechanical, thermal, and chemical
stability of this interconnect material are critical factors controlling
the ability to fabricate high voltage series-connected cells along the
length of the support tube. Accurate matching of component coefficients
of thermal expansion is necessary to prevent cracking and reactive
mixing during initial startup heating and subsequent thermal cycling.
Apparently, cell fabrication procedures also influence stability through
their effect on final component morphology and microstructure. Addi-
20
tional constraints on the critical interconnect section include:
o Moderate material cost.
o Invariant composition in both air and fuel atmospheres.
o No deleterious reaction with other cell components at
1,000°C (1,800°F).
o Resistivity less than 20-50 ohm cm and nearly 100% electronic
conduction at 1,000°C (1,800°F).
o Negligible metal ion conduction.
Westinghouse, the principal investigator of solid oxide technology
in the United States, has made substantial progress in this area in
recent years, although considerable effort remains to demonstrate viable
21
large-scale structures. As usual, materials problems dominate at
high temperature. Problems involving the migration of dopant ions at
elevated temperature under the influence of the inter-electrode po-
tential, and electrolyte "aging" or ionic conductivity degradation due
09
to a reordering of atomic defects, have been discussed. Possible
electronic (in addition to ionic) conductivity of the solid oxide
11-35
-------
electrolyte may further complicate the progress of solid oxide tech-
nology.
Systems studies on large, coal-fueled, central-station power plants
provide the incentive for further development of solid oxide technology.
Q
Initial studies by Westinghouse in 1970 and recently, as part of the
22
NASA-sponsored Energy Conversion Alternatives Study, suggest that
coal-to-busbar efficiencies exceeding 50% might be achieved. Similar
studies by General Electric indicate lower efficiencies, and much re-
mains to be clarified concerning system installed cost.
As with the molten carbonate system, the high-temperature waste
heat from the solid oxide cell can be recovered effectively as steam for
use in a bottoming cycle. On the other hand, little is known of the
effect of impurities (H^S, COS, particulates) in the gasified coal
feed on solid oxide cell performance and life. As stated earlier, this
technology appears limited to use in large-scale continuously operated
power plants rather than smaller cycling plants.
4. Alkaline Fuel Cells
Alkaline electrolytes (e.g., KOH) may also be used to provide the
ionic pathway between fuel and air electrodes. In particular, the
oxygen reduction reaction appears to proceed more rapidly at high pH,
compared with acid (low pH) electrolytes. Further, the somewhat less
corrosive properties of aqueous alkaline electrolytes opens up the
possibility of finding less expensive, stable, nonnoble metal cata-
lysts. The characteristic electrochemical reactions in alkaline fuel
calls are:
Anode reaction: H2 + 20H~
Cathode reaction: 1/2 02 + 1^0 + 2e"
Overall reaction: H2 + 1/2 02 • H20 + direct
electrical energy
11-36
-------
Alkaline fuel cells have reached an advanced state of development
for aerospace applications where high efficiency (hence, low reactant
23
payload requirements) is a premium. Such systems use ultrapure hy-
drogen and oxygen reactants. Attempts to develop a terrestrial fuel-
cell system based on alkaline electrolyte technology were recently
carried out by Exxon, using thin, inexpensive plastic fuel-cell
components developed jointly with the Alsthom Company in France.
The principal difficulty facing alkaline fuel cells in terrestrial
applications involving carbonaceous fuels (hydrocarbons, coal, methanol)
is lack of electrolyte invar iance in the presence of carbon oxides:
20H
Effective procedures for removing carbonate ion, regenerating the spent
caustic electrolyte, or preventing carbon oxides from entering the fuel
cell must be developed. Novel electrochemical techniques were explored
for the former, but considerable effort is required for development to
commercial status. On the other hand, adequate existing technology is
available for precleaning fuel gases containing C0? (and H_S) and
for removing trace amounts of CO. from incoming ambient air (350 ppm
CO.). These processes include absorption in promoted carbonate or
caustic solution, and pressure- swing adsorption cycles using molecular-
sieve adsorbents. Further, the efficiency and cost of these pretreat-
ment steps appear reasonable. Large, 400-MW, intermediate-load
power plants were designed, using reformed naphtha or methanol fuel,
with heat rates between 7,100 and 7,900 kJ/kWh (6,700 and 7,500
Btu/kWh). These values compare favorably with those of other second-
generation technologies. These low heat rates were projected, although
the assumed temperatures for the alkaline system were between 70 and
120°C (158 and 148°F), too low for effective conversion of waste
heat to electricity via a bottoming cycle. Rather optimistic levels of
cell performance were used, assuming a successful but costly R&D program
to overcome problems with internal component contact resistance and to
develop improved cathode electrocatalysts. Prospects exist for
11-37
-------
substituting spinel and perovskite oxide cathode catalysts for platinum,
but these remain to be explored fully.
Although alkaline fuel-cell technology is not being developed for
utility power generation at present, it remains the most promising
approach for electricity production from relatively pure hydrogen fuels
that may be available from future thermochemical or electrolytic hydro-
gen production. Indeed, the operating characteristics of the current
second-generation molten carbonate technology are such that CC^ would
actually have to be manufactured independently to maintain electrolyte
invariance in such applications.
5. Ion Exchange Membrane Cells
The ion exchange membrane fuel cell uses a hydrogen ion-conducting
solid polymer as electrolyte. Initial ion exchange membranes were
formed using the strong acid, polystyrene divinyl benzene sulfonic acid,
although recently membranes have been formed from more stable and costly
fluorinated polymer resins. Electrochemical reactions in the ion ex-
change membrane cell are similar to those noted earlier for the phos-
phoric acid system.
Ion exchange membrane calls have been under continuous development
by General Electric since the early 1960s, primarily for aerospace
7 S
applications such as the Gemini missions. From time to time, at-
tempts have been made to exploit this technology in terrestrial power
26
plants. Initial problems with dimensional and thermal stability
have been overcome, resulting in an extension of cell operating tempera-
tures to somewhat more than 100°C (212°F). Care must be exercised
to prevent membrane drying, particularly when dilute reactants (reformed
fuels, air) are used.
The lower operating temperature of the ion exchange membrane,
coupled with its acidic nature, appear to be the principal technology
11-38
-------
issues limiting its widespread application. Noble metal catalysts are
required and the prospects for raising catalyst utilization by operating
at much higher temperature (200°C or 390°F) are restricted. Fur-
ther, catalyst sensitivity to the trace CO content of reformed fuel
feeds may require the use of alloy anode catalysts. Lastly, opportun-
ities for effective waste heat recovery are also limited by the low
operating temperature, as with current alkaline fuel-cell technology.
11-39
-------
REFERENCES
1. Bolan, P., "Heat Pumps and Fuel Cells," Paper 23d, 69th Annual
AIChE Meeting (November 30, 1976)
2. United Technologies Corporation, Power Systems Division, "Venture
Analysis Case Study for On-Site Fuel Cell Energy Systems,"
Report No. FCR-0787, final report in three volumes (July 31,
1978).
3. Stickles, R. P., et al., "Assessment of Industrial Applications for
On-Site Fuel-Cell Cogeneration Systems," NASA CR-135429
(September 1978).
4. United Technologies Corporation, Power Systems Division, "National
Benefits Associated with Commercial Application of Fuel-Cell
Power Plants," ERDA 76-54, UC-93 (February 17, 1976).
5. Public Service Electric and Gas Company, "Economic Assessment of
the Utilization of Fuel Cells in Electric Utility Systems,"
Final Report, EPRI EM-336 (November 1976).
6. Gillis, E. A., "Fuel Cells for Dispersed Generation of Electric
Power," International Conference on Energy Use Management,
Tucson, Arizona (October 24, 1977).
7. Burns and McDonnell Engineering Company, "An Assessment of the Fuel
Cell's Role in Small Utilities," EPRI Report EM-696, Vol. 1
(February, 1978).
8. King, J. M., "Integrated Coal Gasifier/Molten Carbonate Fuel-Cell
Power Plant Conceptual Design and Implementation Assessment,"
NASA CR-134955, FCR-0157 (October 19, 1976).
9. Sverdrup, E. G., et al., "Project Fuel Cell," Final R&D Report No.
57, Westinghouse (1970).
10. Inside D.O.E., p. 2 (November 27, 1978).
11. Fickett, A., Fuel Cell R&D Status Reports, EPRI Journal, p. 14
(April 1976), p. 46 (June/July 1977), p. 34 (June 1978).
12. Pan, Y. C., et al., "Fuel Cell Catalyst Sintering Studies," Final
Report, EPRI EM-833 (July 1978).
13. Communications from Dr. P. Stonehart, Stonehart Associates.
14. Communications from Professor E. Yeager, Case Western Reserve
University, Cleveland, Ohio.
11-40
-------
15. Blurton, K. F., et al., "The Performance of Molten Carbonate Fuel
Cells," presented at the llth international Power Sources
Symposium, Brighton, England (1978).
16. Kunz, M. R., "Molten Carbonate Fuel-Cell Performance," National
Fuel Cell Seminar, San Francisco, California, July 11-13, 1978,
Abstracts, p. 96.
17. King, J. M., "Advanced Technology Fuel-Cell Program," Interim
Report, EPRI EM-576 (November 1977).
18. Molten Carbonate Fuel-Cell Workshop, held at Oak Ridge National
Laboratories, October 31-November 2, 1978.
19. Weaver, R. D., et al., "Direct/Coal Air Fuel Cell," National
Fuel-Cell Seminar, Boston, Massachusetts, June 21-23, 1977,
Abstracts, p. 77.
20. Ruka, R. J., et al., "Electrical Interconnection of High
Temperature, Solid Oxide Fuel Cells," National Fuel-Cell
Seminar, Boston, Massachusetts, June 21-23, 1977, Abstracts, p.
79.
21. Ruka, R. J., "Development of a Practical Interconnection for Solid
State High Temperature Fuel Cells," National Fuel-Cell Seminar,
San Francisco, California, July 11-13, 1978, Abstracts, p. 159.
22. Dragoo, A. L., et al., "Solid Oxide Fuel Cells," National Fuel-Cell
Seminar, Boston, Massachusetts, June 21-23, 1977, Abstracts, p.
91.
23. Bolan, P., et al., "Hydrogen-Oxygen Alkaline Fuel Cells," National
Fuel-Cell Seminar, Boston, Massachusetts, June 21-23, 1977,
Abstracts, p. 115.
24. Elzinga, E. R., et al., "Application of the Alsthorn/Exxon Alkaline
Fuel Cell System to Utility Power Generation," Final Report,
EPRI EM-384 (January 1977).
25. McElroy, J. F., "Status of H2/02 Solid Polymer Electrolyte
Fuel-Cell Technology," National Fuel-Cell Seminar, Boston,
Massachusetts, June 21-23, 1977, Abstracts, p. 95.
26. MacLeod, E. N., et al., "Evaluation and Optimization of Solid
Polymer Electrolyte (SPE) Fuel Cells," Final Report, MPR-022
(May 15, 1978).
11-41
-------
III. SELECTION OF ENERGY SUPPLY SYSTEMS FOR DETAILED ANALYSIS
A. Criteria for Proposed Systems
To analyze five energy supply systems in detail, we were initially
to propose 10-15 energy systems, of which several were to utilize fuel
cells. To put all proposed systems on a common basis, EPA specified the
following constraints:
o All systems will utilize western coal.
o All systems will provide residential heating and cooling.
o All systems will be based on advanced technology capable of
being commercialized in the period 1980-2000.
o The residences to be provided heating and cooling will be
located 800-1,600 km (500-1,000 mi) from the coal mine.
In addition to the constraints listed above, some other limiting
factors had to be applied to the selection of systems to ensure that
they were representative of the array of technologies that would actu-
ally be available to utilities in the coming decades, and to which they
would give serious consideration. We only examined systems that could
readily supply the type of load specified. For example, heating and
cooling tend to constitute a large part of a utility's intermediate- and
peak-load demand. Thus, intermediate and peaking electricity generation
technologies are more appropriate than base-load technologies for
inclusion in the proposed systems.
Market penetration studies have indicated that the primary applica-
tion of fuel cells in future utility generation mixes will be as inter-
mediate-load devices. This result is supported by two simple econ-
omic facts: (1) the high cost of fuels, especially coal-based fuels,
III-l
-------
that fuel cells must use makes them too expensive for base-load use; and
(2) gas turbines will almost always be cheaper to use as peak-load
devices because of their lower capital cost.
Thus, to make all proposed systems under consideration consistent
with likely fuel-cell systems, we limited the electricity generation
components of the systems to intermediate-load devices. This choice
eliminated those technologies which, because of high capital cost or
lack of cycling capability, are primarily restricted to base-load appli-
cation. For example, fluidized bed combustion, coal gasification
coupled with either a combined-cycle system or a molten carbonate fuel
cell, and conventional coal-fired steam technology were not included.
The exclusion of the latter technology is limited only to new plants,
however, because older base-load plants are commonly reassigned to
intermediate-load service. Naturally, technologies such as gas turbines
that are primarily used in peak-load applications were also excluded.
Finally, those technologies that, although very promising and with
some likelihood of achieving commercialization by the year 2000, but
that are not at a stage of development sufficiently advanced to enable a
thorough analysis, were not considered. An example of such a technology
is the direct coal-fueled fuel cell.
To further aid in defining the systems choices, we chose a
geographical setting for the end use. To be generally applicable, this
location should have climate characteristics typical of regions of the
United States where a significant fraction of the population lives. The
most populated area of the country is in the northeastern corridor.
That climatic zone has a winter heating season of 2,800 - 3,900 Centri-
grade degree-days (5,000-7,000 Farenheit degree-days) and a summer
cooling season of 280-830 Centigrade degree-days (500-1,500 Farenheit
degree days) .
&
A degree-day is the difference between the average temperature and
18°C (65°F) over one 24-hour period.
III-2
-------
Because that region is too far from the western coal fields for
economical use of western coal or its conversion products, a more
westerly location was considered desirable for analytical purposes. The
region bounded by Kansas City, Des Moines, and Omaha appeared to be
suitable. Its climatic conditions of 3,300 heating and 560 cooling
Centigrade degree-days (6,000 heating and 1,000 Farenheit cooling
degree-days) are similar to those of the heavily populated Northeast,
and its distance from the coal fields of Wyoming and Montana is about
1,300 railroad kilometers (800 miles). Thus, this region was chosen as
the location for the end-use components of all the systems considered.
B. Proposed Systems
Twelve energy supply systems that met all the criteria previously
set forth were proposed for evaluation. Because of similarities in fuel
type, energy conversion technology, and end-use, these systems were
classified into four broad system types for supplying residential
energy, each having one or more fuel supply options (see Table III-l).
A brief discussion of each type, including a system diagram, follows.
1. Type 1; Conventional Power Plant/SNG
Type la is a dual system in which residential cooling is provided
by an air conditioner, and heating is supplied by a gas furnace (see
Figure III-l). The electricity that powers the air conditioner is
generated by an older coal-fired power plant that has been reassigned to
intermediate-load service. It is equipped with a stack gas scrubber to
remove S02. The gas furnace burns synthetic natural gas (SNG) de-
rived from coal. The power plant is assumed to be located near the
point of electricity use, requiring the coal to be shipped by unit train
from the mine. The SNG facility is assumed to be located near the mine,
requiring the transport of gas via interstate pipelines.
This system type was chosen to represent, as closely as
possible, the conventional energy supply system in the West North
III-3
-------
Table III-l
CATEGORIZATION OF PROPOSED RESIDENTIAL ENERGY SUPPLY SYSTEMS
Off-site Generation of Electricity
Electric/Gas Residence
Type la: Coal-fired power plant; SNG
Type Ib: Coal-derived oil-fired power plant; SNG
All Electric Residence
Type 2a: 26-MW fuel cell; methanol
Type 2b: 26-MW fuel cell; SNG
Type 2c: 26-MW fuel cell; coal-derived naphtha
Type 2d: 26-MW fuel cell; hydrogen
Type 3a: Combined-cycle power plant; methanol
Type 3b: Combined-cycle power plant; SNG
Type 3c: Combined-cycle power plant; coal-derived
distillate fuel
On-Site Cogeneration of Electricity and Heat
All-Electric Residence
Type 4a: 100-kW fuel cell; methanol
Type 4b: 100-kW fuel cell; SNG
Type 4c: 100-kW fuel cell; coal-derived naphtha
III-4
-------
•COAL MINE
UNIT TRAIN
COAL-FIRED POWER PLANT
ELECTRICITY TRANSMISSION
AND DISTRIBUTION
AIR CONDITIONER
COAL GASIFICATION PLANT
GAS PIPELINE
GAS DISTRIBUTION
GAS FURNACE
FIGURE 111-1. TYPE 1a: COAL-FIRED POWER PLANT/SNG
III-5
-------
Central region of the United States, which includes Kansas City, Des
Moines, and Omaha. Of course, no SNG is being used at this time, but it
would be a likely candidate as a supplemental source of pipeline gas in
the future when natural gas is expected to be in short supply. Elec-
tricity generation in that part of the country is largely carried out in
direct coal-fired steam plants. Older plants are commonly reassigned
from base-load to intermediate-load service, with peak-load service
typically provided by gas turbines. The use of stack gas scrubbers on
plants burning low-sulfur western coal is likely to be required to meet
revised federal new source performance standards (NSPS)-
Type Ib was proposed as an alternative to Type la in which a
coal-derived liquid fuel would be burned in an oil-fired power plant to
provide electricity for air conditioners (see Figure III-2). This sys-
tem type would require the construction of new oil-fired power plants
because very few are now in operation in the West North Central states.
The main advantage of this system type, compared to Type la,
is environmental, because clean fuels would be burned in place of coal.
The resulting cost of electricity would be very high, however, because
of high fuel costs and high capital recovery costs resulting from the
construction of new plants that would operate with a load factor of only
30-40%.
The option of producing solvent-refined coal (SRC) for Type Ib
was initially considered. However, for low-sulfur western coal, very
little benefit is achieved relative to the high processing costs. Up-
grading the product to a liquid fuel that can be sent through pipelines
makes more sense, because transporting fuel by pipeline saves enormously
on shipping costs and a liquid fuel would have wider market potential.
III-6
-------
COAL MINE
COAL LIQUEFACTION
PLANT
FUEL OIL
PIPELINE
OIL-FIRED
POWER PLANT
ELECTRICITY TRANSMISSION
AND DISTRIBUTION
AIR CONDITIONER
COAL GASIFICATION
PLANT
GAS PIPELINE
GAS DISTRIBUTION
GAS FURNACE
FIGURE 111-2. TYPE 1b: OIL-FIRED POWER PLANT/SNG
III-7
-------
2. Type 2; 26-MW Fuel-Cell Power Plant
The 26-MW fuel cell was chosen for Type 2 as representative of
the size envisioned for near-term application of first generation fuel
cells. A 26-MW device (which actually consists of six modules of about
4.5 MW each) is small enough to be located at dispersed sites throughout
a utility grid (see Figure III-3). Furthermore, units can be added
incrementally to meet increasing loads, thus deferring large com-
mitments of capital required for larger units. Because fuel cells are
environmentally unobtrusive at the point of operation they are easily
sited, even in urban areas.
Coal-based fuels can be supplied to fuel cells in several
ways. Because the small size and dispersed location of the fuel cells
precludes on-site coal conversion activities, location of coal con-
version facilities at the mine was assumed. Four fuels — methanol
(Type 2a), SNG consisting primarily of methane (Type 2b), coal-derived
naphtha (Type 2c), and hydrogen (Type 2d) — are compatible with fuel-
cell operation, and can be transported over long distances by pipeline.
All the fuels except hydrogen must be reformed at the fuel-cell site.
SNG and hydrogen could be distributed to the fuel cells by pipeline,
while liquid fuels would be distributed by trucks. On-site storage
capability would be required for the liquid fuels.
The electricity produced by the fuel cells is assumed to power
heat pumps located in individual residences. The heat pumps supply
heating in the winter, and are operated as air conditioners in the
summer.
The use of heat pumps implies that the residences are all-
electric. Although this does not typically occur in the region under
consideration, it is an option when natural gas may be in short supply.
Already in many areas of the country obtaining natural gas connections
for new homes is not possible. In such cases, all-electric homes equip-
ped with heat pumps provide a viable, reasonably economical alternative.
III-8
-------
I
VO
(A)
COAL-TO-METHANOL
PLANT
METHANOL PIPELINE
METHANOL STORAGE/
DISTRIBUTION
COAL MINE
(B)
COAL GASIFICATION
PLANT
NATURAL GAS PIPELINE
NATURAL GAS
DISTRIBUTION
(C)
COAL LIQUEFACTION
PLANT
NAPHTHA
PIPELINE
NAPHTHA STORAGE/
DISTRIBUTION
26-MW FUEL CELL -*•
ELECTRICITY DISTRIBUTION
HEAT PUMP
CO/>
HY
FIGURE 111-3. TYPE 2: 26-MW FUEL CELL
-------
In addition, the high electrical demand implied by the con-
struction of new all-electric homes necessitates new electrical gener-
ation capacity of all kinds. Not enough older fossil steam plants are
likely to be available to reassign to the intermediate use for which
fuel cells are well suited; thus, the construction of new facilities is
required.
3. Type 3; Combined-Cycle Power Plant
Type 3 is proposed as an alternative to fuel cells that can
economically provide intermediate electrical loads (see Figure III-4).
The low capital costs and low heat rate of combined-cycle systems make
them competitive with fuel cells. The end use for electricity is id-
entical to that in Type 2. The system type envisioned here would con-
sist of a gas turbine coupled with a steam bottoming cycle. Suitable
coal-derived fuels for the combined-cycle plant include SNG (Type 3a),
distillate fuel (Type 3b), and methanol (Type 3c). (A hydrogen system
was not proposed because without the advantage of avoiding reforming
there are no benefits to balance the additional costs and problems of
hydrogen transport and handling.) Combined-cycle power plants would be
sized in the 200-300 MW range so that dispersed siting would not be
practical. However, the fuel distribution costs would be reduced
compared to the fuel-cell case.
4. Type 4; 100-kW Fuel-Cell Power Plant with Heat Recovery
Type 4 takes advantage of the ability to construct small fuel-
cell power plants so they can be located at the site of electricity
demand (see Figure III-5). If this site is an apartment building or
cluster housing complex, the fuel-cell waste heat that is normally
rejected to the atmosphere may be recovered and used to supply domestic
hot water and supplemental heat for heat pumps.
111-10
-------
•COAL MINE
(A)
COAL-TO-METHANOL
PLANT
METHANOL
PIPELINE
METHANOL
DISTRIBUTION
(B)
COAL GASIFICATION
PLANT
GAS PIPELINE
GAS DISTRIBUTION
COMBINED CYCLE
POWER PLANT
ELECTRICITY TRANSMISSION
AND DISTRIBUTION
HEAT PUMP
(0
COAL LIQUEFACTION
PLANT
FUEL OIL
PIPELINE
FUEL OIL
DISTRIBUTION
FIGURE 111-4. TYPE 3: COMBINED-CYCLE POWER PLANT
III-ll
-------
COAL MINE
(A)
COAL-TO-METHANOL
PLANT
(B)
COAL GASIFICATION
PLANT
(C)
COAL LIQUEFACTION
PLANT
METHANOL
PIPELINE
GAS PIPELINE
METHANOL
STOR AG E/DISTRIBUTION
NAPHTHA
PIPELINE
GAS DISTRIBUTION
NAPHTHA
STORAGE/DISTRIBUTION
100-kW FUEL CELL
WITH HEAT RECOVERY
HEAT PUMP
FIGURE III-5. TYPE 4: 100-kW FUEL CELL WITH HEAT RECOVERY
111-12
-------
The fuel supply options for this system are the same as for
Type 2, except that hydrogen is omitted, because of anticipated dif-
ficulty of installing a dense hydrogen distribution network in the time
frame considered by the study. Thus, suitable coal-derived fuels are
methanol (Type 4a), SNG (Type 4b), and naphtha (Type 4c).
C. Evaluation of System Components
The next step in this study was to select five energy-supply
systems for detailed analysis from the 12 systems initially proposed.
The selection was based on preliminary estimates of cost, energy effi-
ciency, and environmental impact. The analyses of the systems that
supported the selection procedure are described below.
1. Cost and Efficiency
The individual system components or process steps for each of
the 12 candidate energy systems were analyzed. Component cost and
energy efficiency were estimated, with data taken from SRI studies or
from the published literature.
The cost and thermal efficiency estimates were prepared from
specific cost studies and were for new equipment. The costs were nor-
malized on the basis of the higher heating value (HHV) of the principal
product. Normalizing costs in this manner avoided analytical anomalies
caused by mismatches in the various plant sizes. All plants were as-
sumed to be sufficiently large so that maximum reasonable advantage
could be claimed for economies of scale. For ease of handling, cost
estimates were compiled from the three major cost components: plant
investment, plant operating costs (exclusive of plant amortization), and
cost of feedstock consumed in the conversion process. A capital recov-
ery factor of 18.2% per year was assumed (equivalent to a regulated
utility cost basis of 65% debt at 10% interest, and 35% equity at 15%
return on equity).
111-13
-------
2. Sulfur Dioxide Emissions
To provide some guidance to the environmental impact of each
energy supply system, we chose a single parameter for analysis. This
parameter was the total amount of SO- emitted by the components of
each system. Because SCL is a major pollutant of concern and is sub-
ject to a wide range of regulatory activity, it provided a convenient
focus for a preliminary environmental analysis. Because the primary
sources of SO- emissions are the coal conversion technologies, the
sulfur balances for those technologies were examined in detail.
Coal conversion has two principal sources of sulfur emis-
sion—the coal-burning utility part of the plant, which burns coal to
provide steam, electricity, and process heat, and the gasification or
liquefaction part of the plant. The utility sulfur emissions are in the
stack gases and consist almost entirely of S0? from oxidation of the
sulfur in the coal. The liquefaction or gasification of coal produces
principally H S in product or by-product gas streams. Small amounts
of carbonyl sulfide and other volatile sulfur compounds may also be
formed. These gas streams are purified by scrubbing systems that use
amines (MEA, DEA) or organic solvents such as methanol. This scrubbing
generates a waste stream of concentrated sulfur compounds which can then
be treated in a sulfur recovery process such as the Glaus process.
Minor amounts of sulfur are contained in ash and unreacted
char streams, as well as in water effluents, but this sulfur is not
included in the present discussion because it has little effect on the
system analysis.
There are three possible strategies for controlling sulfur
emissions:
Each installation would be controlled to achieve some standard
emission rate per GJ (0.948 million Btu) of coal input (e.g.,
NSPS). The major impact on particular technologies would be
cost differences, because for some processes the capital re-
quired to meet the standard will be higher than for others.
111-14
-------
o Each installation would have the same total expenditure for
air pollution control. The major impact on particular tech-
nologies would be differences in sulfur emission rates. For
some conversion technologies this fixed expenditure would be
sufficient to achieve very low sulfur emission rates whereas
for others it would not.
o Each installation would have the best available technology
installed. Different coal conversion technologies would have
different control costs and different sulfur emission rates
with this strategy.
The third option was chosen for this study; the best available
technology was defined to be:
o Electrostatic precipitators and lime/limestone scrubbers for
stacks of coal-fired boilers. Scrubbing removes 85% of the
stack gas sulfur.
o Glaus or Stretford units followed by incinerators and lime/
limestone scrubbers used on sulfur-rich gases leaving the
conversion process. The Glaus or Stretford unit removes
90-95% of the feed sulfur, and the incinerator and scrubber
removes 85% of the sulfur remaining in the tail gas.
Those particular choices of technology resulted in sulfur
emissions that are somewhat lower than the proposed federal NSPS for
coal gasification and also yielded boiler stack gases that are cleaner
than required by current federal and state standards for new sources.
The technology chosen will also be suitable for controlling carbonyl
sulfide, hydrocarbons, particulates, as well as other pollutants that
will be considered later in this study.
The control technology is fixed so that each coal conversion
technology will have different sulfur emissions, primarily because of
differences in process efficiency. Low efficiency technologies such as
methanol production have larger utility requirements, resulting in a
larger percentage of the coal delivered to the plant going to the util-
ity plant rather than to the gasification or liquefaction reactor. With
the choices of best available technology made for this study, coal
burned in the utility plant releases 15% of its sulfur to the environ-
111-15
-------
ment whereas coal processed in the converter releases less than 1% of
its sulfur to the environment.
3. Ranking of Proposed Systems
Using the parameters estimated for each system component, the
overall costs, efficiency, and release of S0« associated with pro-
viding residential heating and cooling were estimated. All quantities
were calculated on the basis of providing one year's heating and cooling
to a single residence. Those figures were then used to rank the twelve
energy systems and aid in the selection of five systems for further
analysis.
Because the estimates of cost, efficiency, and SO. emissions
for both the system components and the total systems were preliminary in
nature, and to avoid confusion with more precise estimates given later
in this report, results of those estimates will not be presented here.
However, the relative rankings of the systems, based on those prelim-
inary estimates, are given in Table III-2.
D. Selection of Systems
Selecting five systems from the twelve proposed was complicated
because multiple parameters in addition to nonquantifiable consid-
erations had to be used in evaluating the systems. In general, a unique
set of systems could not be chosen without attaching precise weighting
factors to each parameter of interest (cost, efficiency, SO- emis-
sions) to obtain an overall impact parameter for each system. Such
impact parameters, tempered by nonquantitative considerations, could
then be used to guide the systems selection. The task was made easier,
however, by recalling that the purpose of the systems selection was to
select systems to evaluate that would be competitive in similar
applications and would represent realistic choices for utilities in the
111-16
-------
Table III-2
RANKING OF COST, EFFICIENCY, AND
EMISSIONS FOR ENERGY SUPPLY SYSTEMS
(lowest cost = 1) (most efficient =1)
la
Ib
2a
2b
2c
2d
3a
3b
3c
4a
4b
4c
1
2
12
7
11
8 (tie)
10
4
5
8 (tie)
3
6
3
6
12
9
10
5
11
7 (tie)
7 (tie)
4
1
2
— z «•
(lowest emissions =
9
7
11
5
5
3
10
4
12
7
1
2
(tie)
(tie)
(tie)
(tie)
111-17
-------
1980-2000 time frame. Each system had to pass judgments as to its
likelihood of implementation in that time frame before other parameters
were weighed. Once that criterion had been satisfied, the cost, effi-
ciency, and SO. emission comparisons could then be applied to the
remaining systems.
Because four broad types of systems with different fuel options
were proposed, we selected at least one system within each type to make
the study representative. Table III-2, which shows the relative ranking
of costs, efficiency and SO. emissions for all the systems was used to
assist in the selection process.
Type la (Figure III-l) was found to have the lowest cost of
heating and cooling of any system, moderately high efficiency, and S02
emissions that are high relative to most other systems. This system
type is closest to current practice in the West North Central region,
and thus represents a case with high likelihood for future use.
The total cost of heating and cooling from Type lb (Figure
III-2) was only slightly higher than that of Type la, even though the
cost of electricity is more than twice as high, because of the rela-
tively small amount of electricity used compared to SNG. S09 emis-
sions were lower and the efficiency was about the same.
At present, there are essentially no large oil-fired steam
electric plants in the West North Central region, and it is questionable
whether any will be built in the future, especially for intermediate
load applications.
In Type 2 (Figure III-3), which included four fuel supply
options, the methanol option appeared to be the least favorable in terms
of cost, efficiency and S02 emissions (primarily because of the high
cost and low efficiency of the coal-to-methanol conversion step), whereas
hydrogen was the most favorable. The SNG and naphtha options were
intermediate. Although hydrogen appeared most attractive in terms of
111-18
-------
these quantitative parameters, several aspects of this system make it
unlikely that it would be used before 2000. The main problem is the
necessity of constructing hydrogen pipelines and distribution systems in
the absence of other significant demands for hydrogen. Because nearly
any use that can be foreseen for hydrogen over the next 25 years could
be filled equally well by methane, there seems little incentive for
constructing hydrogen systems where natural gas systems already exist.
Furthermore, some serious technical problems, such as embrittlement of
pipeline steel, would have to be overcome before hydrogen systems could
be constructed. In addition, no one is likely to construct a hydrogen
system solely dedicated to supplying fuel cells, whereas the presence of
existing natural gas and petroleum distribution systems favors the use
of SNG, naphtha, or methanol.
In Type 3 (Figure III-4), methanol again ranked unfavorably,
having the highest cost, lowest efficiency, and high SO emissions
relative to the other fuels. SNG and fuel oil rank approximately
equally in cost and efficiency. The higher SO emissions from the
fuel oil option result from the small amount of sulfur in the fuel oil
that is released during combustion. In the future, utilities may be
restricted in their use of natural gas for electricity generation. Cer-
tainly, natural gas will be phased out as a boiler fuel, but whether it
will continue to be allowed as turbine fuel is not clear. SNG will
probably be subject to the same restrictions as natural gas because it
will serve the same needs, and the tendency will be to preserve it for
higher priority users (e.g., residential and commercial customers).
In Type 4 (Figure III-5) cost, efficiency, and S09 emissions
are improved compared to Type 2 because of the recovery of waste heat
from the fuel cell. Among the three fuel options, methanol appeared
the least attractive overall for the reasons mentioned previously. SNG
is superior in cost and efficiency to naphtha, but comparable in SO
emissions. Assuming that gas or electric utilities were to own and
maintain small fuel cells located in housing complexes, there does not
seem to be an overwhelming advantage to using SNG rather than naphtha,
111-19
-------
except that supplying SNG through the gas distribution grid would be a
more convenient, and perhaps a more secure, source of supply than deliv-
ering naphtha by truck.
The final selection of five systems resulted from several
judgments about their relative attractiveness. First, compared to other
fuels, methanol always appeared less attractive in terms of cost, ef-
ficiency, and SO. emissions. Therefore, Types 2a, 3a, and 4a were
eliminated.
Second, large-scale systems for transporting and distributing
hydrogen are unlikely to be constructed before the end of the century.
Thus, Type 2d was removed as an option.
Third, future restrictions on natural gas use are likely to
limit its availability (SNG included) for large new electricity gen-
eration facilities. Therefore, coal-derived fuel oil would be a
likelier prospect as a fuel for combined-cycle power stations. This
judgment favored option 3c over 3b.
Consideration of the merits of Type Ib led to a fourth judg-
ment: that there will be relatively little incentive to build new oil-
fired steam plants to meet intermediate loads in the West North Central
region. Compared to combined-cycle plants, oil-fired steam plants are
more capital intensive and less efficient, resulting in a 13% higher
cost of electricity when coal-derived fuel oil is used. Thus, Type Ib
was eliminated.
Of the remaining six systems there was no clear-cut single
choice for elimination. We retained Types la and 3c to compare with
the fuel-cell options, however. This left two fuel supply options for
the 26-MW fuel cell (Types 2b and 2c), and two for the 100-kW fuel cell
(Types 4b and 4c). We judged that because the thrust of research to
date on fuel cells for on-site residential applications has been
directed toward the use of natural gas as a fuel, the SNG supply option
111-20
-------
should be retained for the 100-kW system; thus, Type 4c was eliminated.
For the 26-MW system, SNG and coal-derived naphtha appeared to be
equally likely. Therefore, both fuel options were retained.
In summary, the five systems selected for detailed analysis are:
o System 1: A coal-fired power plant that supplies electricity
to residences and a coal gasification plant that supplies SNG;
electricity is used to power air conditioners and SNG is
burned in gas furnaces (Type la).
o System 2: A 26-MW fuel-cell power plant fueled by SNG derived
from coal; electricity is provided to residences that are
heated and cooled by heat pumps (Type 2b).
o System 3: 26-MW fuel-cell power plant fueled by coal-derived
naphtha; electricity is provided to residences that are heated
and cooled by heat pumps (Type 2c).
o System 4: A combined-cycle power plant fueled by coal-derived
fuel oil; electricity is provided to residences that are
heated and cooled by heat pumps (Type 3c).
o System 5: A 100-kW fuel-cell power plant fueled by SNG, sited
in a housing complex; electricity is provided to residences
that are heated and cooled by heat pumps. In addition, heat
is recovered from the fuel cell to supply supplemental space
heating and hot water (Type 4b).
111-21
-------
E. References — Chapter III
1. W. Wood, M. P. Bhavaraju, and P. Yatcko, "Economic Assessment of
the Utilization of Fuel Cells in Electric Utility Systems,"
Electric Power Research Institute Report EM-336 (November
1976).
111-22
-------
IV CONCEPTUAL SYSTEMS DESIGNS
After the five energy supply systems were selected for analysis,
the next task was to provide conceptual designs of the systems so that
further economic, environmental, and energy efficiency analysis could be
carried out. The conceptual designs were to specify technology types,
sizes, flow rates, and other characteristics. Those specifications were
subject to limitations on knowledge of the actual process designs that
will characterize commercial-scale facilities built around advanced
energy technologies. All our conceptual designs of advanced energy con-
version processes can be considered representative but not definitive;
other designs may be equally credible.
Generally speaking, the conceptual designs of advanced processes
were based on optimistic extrapolations of current pilot plant data.
Only time will tell whether these results will be borne out in large-
scale commercial facilities and whether major process changes from the
ones now envisioned will be required. Nevertheless, the conceptual de-
signs represent the process features that must be attained if advanced
energy conversion concepts are to achieve a significant market within
the next 20 years.
The conceptual designs presented in the chapter are described in
terms of the functional units or "building blocks" of which each system
is composed. Each building block defines a system component that is
logically sized to achieve economies of scale and that represents cur-
rent or anticipated future industrial practices. Thus, the components
are not sized to represent equal flow of energy; the output of energy
from one component in a system does not equal the energy input to the
following component. One may envision, for example, that the output
from several coal gasification plants would be required to fill a major
interstate pipeline, or that the output of such a pipeline would supply
IV-1
-------
many 26-MW fuel-cell power plants in addition to homes, businesses, and
industries. Thus, the mismatches in size inherent in the systems repre-
sent the situation that would likely prevail if such systems were actu-
ally constructed. Whenever possible in the following descriptions, the
energy output of one component will be related to the energy input re-
quirements of the following component in the system.
The conceptual design of each component consists of a general de-
scription, followed by a more specific description of equipment size,
flow rates, and so on. When the system component is an energy conver-
sion facility (e.g., coal gasification plant, fuel-cell power plant), a
block flow diagram showing major process elements is presented, along
with material balances. The level of detail is sufficient to allow an
accurate determination of costs, energy efficiencies, and pollutant
emissions in subsequent analyses. Excessive detail has been avoided.
In the following descriptions some components will be common to
more than one system. In such cases, the description of the component
will be presented only once, in the discussion of the first system in
which it appears. In subsequent discussions, the reader will be re-
ferred to the initial description of the component.
A. System 1
A block flow diagram of System 1 is shown in Figure IV-1. The com-
ponents of the system are described below.
1. Coal Mine
The mine that supplies coal for electricity generation, gasif-
ication, and liquefaction is a large surface mine typical of those oper-
ating and planned in the Powder River Basin of northeastern Wyoming and
southeastern Montana. Such mines extract low-sulfur subbituminous coal
IV-2
-------
Figure IV-1
COAL MINE
UNIT TRAIN
COAL
GASIFICATION
PLANT
COAL-FIRED
POWER PLANT
GAS PIPELINE
ELECTRICITY
DISTRIBUTION
GAS
DISTRIBUTION
GAS FURNACE
AND AIR
CONDITIONER
FIGURE IV-1. BLOCK FLOW DIAGRAM OF SYSTEM 1
IV-3
-------
from very thick, relatively shallow seams. Because mining operations
vary according to the geological conditions associated with particular
mining sites, we have selected a hypothetical mining operation that re-
presents more or less typical conditions.
The hypothetical mine produces 4.5 million tonnes (5 million
tons) of coal per year from a seam 15 m (50 ft) thick. The average
overburden thickness is 21 m (70 ft). The characteristics of the coal
produced from the mine, based on Rosebud Seam coal mined in Rosebud
County, Montana, are as follows:
Proximate Analysis:
Moisture 22.0%
Volatile matter 29.4
Fixed carbon 42.6
Ash 6.0
Ultimate Analysis (Dry):
Carbon 67.7%
Hydrogen 4.6
Nitrogen 0.85
Oxygen 18.5
Sulfur 0.66
Ash 7.7
Heating Value: 20.4 MJ/kg (8800 Btu/lb)
Surface mining of 4.5 million tonnes of coal per year requires
large amounts of accessible coal, large equipment to remove the earth
above the coal, and large trucks to haul the coal away. Surface mining
can be accomplished in several ways — using large power shovels, drag-
lines, or bucketwheel excavators. We assume that the dragline method is
used, as illustrated in Figure IV-2.
The procedures involved in operating a large surface mine are
extensive. After permits have been acquired and the coal seam has been
mapped, the base camp office and equipment maintenance building and yard
IV-4
-------
BENCH
Source: Reference 1
FIGURE IV-2. DRAGLINE METHOD OF OVERBURDEN REMOVAL
-------
are set up. Storage areas for spare parts and explosives are located,
access roads to the mining area are established, and electric power for
the office and for the dragline operation are provided. The first exca-
vation removes the ground cover and topsoil. The soil is stockpiled and
later replaced on the mined-over areas to help reestablish ground
cover. If overburden under the topsoil and above the coal cannot
support plant life, it has to be segregated. The earth moving equipment
(such as bulldozers and scrapers) used for removing the soil is also
used in the construction and maintenance of the access roads.
Next, a rotary-type rig drills blasting holes from 14 cm (5.5
in.) to 39 cm (15.5 in.) in diameter in a grid pattern with 15 x 18 m
(50 x 60 ft) spacing down to the top of the coal seam. The explosives
set off in the holes loosen the overburden and make it less difficult to
remove. Draglines then remove the overburden.
The largest draglines today have bucket capacities between 137
3 3
and 158 m (180 and 220 yd ). The boom is as long as a football
field (97 m or 300 ft) and the cab is as tall as a five-story building
and a third as long as the boom. Draglines offer the most versatile way
to remove the overburden, as they are able to remove large amounts with
each pass of the bucket. They can also be used for coal seams with a
variety of depths.
After the dragline removes the overburden, it swings to one
side and drops the overburden in the swath previously cut. Overburden
from the first cut is dropped on the surface. Subsequent cuts fill the
previous hole after the coal is removed. At the end of the stripping
operation the last cut is left open for lack of overburden to put into
the hole. To date, it has not been cost-effective to haul the first cut
overburden pile left on the surface to the remaining hole, but the 1977
Surface Mining Control and Reclamation Act and state laws now require
that the remaining highwall be removed and the area returned to its
original contour as closely as possible. That will require grading the
first pile that was left on the surface and filling in the last cut so
that no highwall remains.
IV-6
-------
The coal is removed first by blasting the seam to loosen the
coal. Rear or bottom dumping trucks are then loaded by a large front-
end loader or power shovel. The coal is hauled from the mine, dumped,
crushed to the proper size, and delivered to a nearby conversion facil-
ity or stored ready for loading on a unit train.
Reclamation must then follow the mining. The overburden is to
be graded as closely as possible to its original contour and the topsoil
replaced. A vegetative cover is established with the goal of achieving
long-term rehabilitation of the mined area.
The following major items of equipment are required to carry
out the mining and reclamation operations:
Quantity Item
1 Dragline, 53 m3 (70 yd3) capacity
1 Power shovel, 12 m3 (16 yd3) capacity
9 Dump trucks, 50-53 m3 (65-70 yd3) capacity
1 Scraper
1 Overburden drill
1 Coal drill
2. Unit Train
In System 1, we assume that unit trains are used to transport
coal from the Powder River Basin coal mine via Gillette, Wyoming, to
coal-fired power plants in the Omaha-Des Moines-Kansas City area, a dis-
tance of approximately 1,300 km (800 mi). We assume for the purposes of
subsequent analysis that 80 km (50 mi) of this distance is newly con-
structed spur line from the coal mine to Gillette, Wyoming. The remain-
ing 1,220 km (750 mi) is previously existing track connecting Gillette
to the tricity area mentioned above.
IV-7
-------
2
Unit trains are a relatively new concept in railroading.
Their principal advantages compared with other railroad services are
lower cost and faster delivery. A large percentage of unit trains in
use today transport coal. They achieve their economies by dedicating an
entire train to continual mass shipments from a single location (the
coal mine) to a single destination (the power plant). Because the en-
tire train has one destinaton, time-consuming stops at train switching
yards can be avoided. The use of the equipment is uniform, so that re-
pairs are more predictable and more easily scheduled.
The train considered in this discussion is new and runs on a
roadbed and rails, parts of which have recently been constructed. The
proportion of newly constructed track is an important characteristic
because it heavily influences the cost of service.
Each train is powered by four 2.24-MW (3,000 horsepower)
diesel-electric locomotives. The coal cars are aluminum-sided light
weight cars with a net carrying capacity of 91 tonnes (100 tons). The
equipment is usually leased by the coal recipient (steam electric power
plant), from the railroad. There is some concern that one hundred
91-tonne (100-ton) cars are too heavy for optimum track and car mainten-
ance cost. However, arguments are advanced that even heavier trains
would be optimal with proper track and car maintenance.
To minimize transportation cost, a railroad seeks to maximize
the amount of coal moved each year. That provides an incentive to mini-
mize train-loading time. Short loading times can be accomplished by
"flood loading," which basically involves building a surge container
over the tracks. The container may be a trackside concrete silo similar
*An effective alternative to the unit train is a coal slurry pipeline.
Generally, the slurry pipeline is more capital-intensive but less labor-
intensive and therefore less vulnerable to inflation than railroads.
Long distances, large quantities, and rugged terrain tend to favor pipe-
lines over railroads. Current discussions on the relative merits of
coal slurry pipelines and unit trains are lively and frequently influ-
enced by political and institutional considerations.3
IV-8
-------
to a grain silo, or a pile of coal on top of a train tunnel. Coal is
dumped into the cars while the train is pulled at 3-6.5 km per hour (2-4
mph) past the coal discharge spout on the surge hopper. Loading rates
of 3,600 tonnes (4,000 tons) per hour have been claimed. Such a
loading rate would allow a train with 9,100 tonne (10,000 ton) capacity
to be filled in 2.5 hours. To be conservative, a full 8-hour shift has
been assumed for loading.
Once a train is loaded, it can proceed directly to its unload-
ing point (the coal-fired power plant). We have assumed an average
train speed (mine-to-mine) of 32 km per hour (20 mph), not including
loading and unloading.
Unloading should be expedited for the same reasons as load-
ing. The fastest unloading method is to pull the train at 6.5 to 10 km
per hour (4 to 6 mph) over an unloading trestle. The trestle has a re-
lease mechanism that opens the bottom of the rapid discharge hopper cars
and dumps their coal between the tracks onto the storage pile beneath
the trestle. Individual cars can be dumped in about 20 seconds. An
entire train can be unloaded in less than 1 hour. As a conservative
estimate for this study, we allowed one 8-hour shift for receiving a
loaded train, unloading it, and sending it back to the mine for refill.
A thawing shed is sometimes necessary at the unloading
station. The thawing shed is used to warm coal in cars that have frozen
during cold weather. Intense radiant heat is used to warm the bottom
and sides of the cars enough that the coal will flow out of the car
during the normal unloading process. Both gas-fired and electric radi-
ant heaters are available. Some electric thawing sheds can thaw five
cars per hour, consuming 5.5 kWh of energy for each tonne of coal
thawed. For the unit train in our example, thawing would take 20
hours and would require 60,000 kWh of electricity.
For the 800-MW coal-fired power plant discussed in the follow-
ing section, approximately 1.3 million tonnes (1.4 million tons) of coal
IV-9
-------
would be required yearly to maintain a load factor of 35%. That re-
quirement could be met by dispatching a 100-car train every 48 hours.
Thus, two unit trains would be dedicated entirely to the operation of
such a power plant.
3. Coal-Fired Power Plant
a. Background
The coal-fired power plant is an older unit that has been
reassigned to intermediate-load service. The nominal unit size is 800
MW. Typically, such units are constructed to provide base-load service
(60-80% load factor). However, after a number of years of base-load
operation, they may be reassigned to intermediate or cycling duty. That
practice is economical for facilities whose capital costs have been par-
tially written off. To employ a new plant in such an application is too
costly because the capital charge rate per kWh would be effectively
doubled with a load factor of only 30-40%.
Because the time frame under consideration in this study
extends to the year 2000, we have assumed that the power plant is built
in the 1970s. Such a plant could be reassigned to intermediate-load
service in the 1990s, or possibly sooner. Consistent with national and
local environmental protection policies, we have assumed that a coal-
fired power plant constructed in the 1970s requires a flue gas desulfur-
ization system as well as a fly ash removal system.
Although flue gas desulfurization (FGD) has been the sub-
ject of intense controversy i.n the past, its use in conjunction with
conventional combustion of coal has become the standard against which
o
alternatives are judged. The principal alternatives now available
for meeting current new source performance standards (NSPS) and state
implementation plans (SIP) are low-sulfur coal and physical coal clean-
ing; the choice depends on site-specific factors. If performance stan-
IV-10
-------
dards are made more stringent or "best available technology" becomes
required, neither low-sulfur coal nor physical coal cleaning alone is
likely to be acceptable. For purposes of this study, lime scrubbing is
chosen as a well-proven technology suitable for treating flue gas from
western subbituminous coal.
b. Process Description
Figure IV-3 is a block flow diagram for a conventional
coal-fired power plant with FGD. The principal stream flows are shown
in Table IV-1 for an 800-MW net capacity, adjusted from those in a re-
cent report according to the coal properties and requirements of the
Q
current study.
In the plant, coal is transported from the coal storage
pile to the raw coal bunkers via conveyor belt. From the bunker it is
fed to the pulverizer where it is ground to about 50 mesh and mixed with
air equal to 10-15% of the total combustion air. The coal/air mixture
is blown to the burners where it is combined with the remaining combus-
tion air (which has been preheated) and combusted. About 20% of the ash
in the coal drops to the bottom of the furnace in solid form, while the
remaining 80% stays with the combustion gases as fly ash.
After combustion, the hot combustion gases pass through
several stages of heat exchange to provide heat to the steam cycle.
These stages consist of: (1) the boiler in which feedwater is evapo-
rated to saturated steam; (2) the superheater, in which steam is heated
beyond the saturation point; (3) the reheater, in which steam exiting
the first stage of the steam turbine is reheated for use in subsequent
stages; and (4) the economizer in which boiler feedwater is preheated.
Modern steam-electric plants typically operate at a superheated steam
pressure and temperature of 24,200 kPa (3,500 psig) and 540°C
(1,000°F), with a 540°C reheat.
IV-11
-------
COAL—*-
M
<
1
10
STORAGE
Ci)
PULVERIZER
AIR
(5) (6)
LIME WATER
1 1
ES
1©
BOILER
~1©
FLY ASH | ^
SLUDGE
STEAM STEAM
^ TURBINE
(9)
1 (3)
~s
BOTTOM
ASH
STACK / S \
GASES \ T \
0 / 1 \
CONDENSER -*—
-800MW
FIGURE IV-3. BLOCK FLOW DIAGRAM FOR AN 800MW COAL-FIRED POWER PLANT
-------
The superheated steam passes through the steam turbines
that power the generator, is cooled in the condenser, and returned to
the boiler. The stack gases, containing fly ash, SO., and other pol-
lutants, are directed first to the electrostatic precipitators (ESPs)
for fly ash removal, and then to the FGD plant for SO- removal, before
being exhausted to the stack. (Operation of the ESP and FGD plant is
described further in Chapter VI.)
The most economical power plant configuration is for a
single furnace/boiler unit to supply steam to a single steam turbine-
generator set. Units of the size required to produce 800-MW net power
output, and larger, are readily available. The steam requirement for
such a plant is about 2.6 x 106 kg/hr (5.7 x 106 Ib/hr).
4. Electricity Transmission and Distribution
Electricity transmission is the high voltage (more than 69
kV), long-distance movement of bulk electrical power. Distribution
is the subdivision of the bulk power from the transmission lines and its
dispersal to the ultimate consumer. Conceptually, transmission and dis-
tribution (T&D) are simple processes; power from a generating station is
sent along a wire to an end user. However, in reality, T&D is very com-
plicated and expensive. Approximately two-thirds of the current cost of
residential electricity is related to T&D expense
Transmission usually includes transforming power at the gener-
ation station up to the voltage of the tranmission line. If a large
amount of power is being moved, very high voltages will allow the most
efficient and least expensive transport (see Figure IV-4). Table IV-2
shows that the highest voltage lines in commercial service in the United
States are 765 kV. However, 1,100- and 1,500-kV systems are under con-
13
sideration. Transmission lines are frequently interconnected to
form large power pools. The redundancy provided by pooling provides a
very reliable power supply to the consumer. However, reliability re-
quirements also contribute to the complexity of planning for new trans-
mission systems.
IV-13
-------
TABLE IV-1
MAIN PROCESS FLOWS FOR 800-MW COAL-FIRED POWER PLANT
1
2
3
4
5
6
7
8
9
Stream
Name and Number
Run-of-Mine Coal
Air
Bottom Ash
Fly Ash
Lime
Water
Sludge
Stack Gas
Superheated Steam
Mass Flow Rate
(103 kg/hr)
410
3,500
5.0
20
4.5
140
26
4,000
2,600
Temperature
°C
16
16
—
150
16
16
49
79
540
(°F)
(60)
(60)
—
(300)
(60)
(60)
(120)
(175)
(1,000)
DC
O
I
I
O
en
UJ
a.
LU
cc
QUANTITY OF POWER TRANSMUTE D-MW
Source: Reference 14
FIGURE IV-4. HIGH-VOLTAGE TRANSMISSION ECONOMY
FOR LONG DISTANCE
IV-14
-------
Table IV-2
ECONOMIC POWER LOADING OF TRANSMISSION LINES
Voltage
(kV)
69
138
230
345
500
765
1,000
Loading
(MW)
15
60
170
430
1,000
2,000
5,000
Source: Reference 13
For the power plant described in the previous section, a
single 765-kV line would be sufficient to transmit the 800 MW of power
generated. However, such an arrangement would hardly ever be encoun-
tered. More likely, two or more 345-kV transmission lines would trans-
mit varying portions of the power to several load centers, or provide
interties to other utility networks.
An example of such an arrangement is the 801-MW Cooper nuclear
plant owned by the Nebraska Public Power System (NEPP). The plant is
connected to utilities serving Omaha, Des Moines, and Kansas City via
three 345-kV lines as well as to other parts of the NEPP grid via a
fourth 345-kV line. The total length of 345-kV lines associated with
the plant is on the order of 480 km (300 mi). A similar arrangement
IV-15
-------
might be expected for a new coal-fired power plant located in the tri-
city region.
Transmission substations, primary distribution lines, distri-
bution substations, and line transformers are included in the increas-
ingly small subdivisions of lower voltage power between the high-voltage
transmission lines and the final customer.
Transmission substations are provided in the grid to inter-
connect different parts of the system. Such a substation may include
high-voltage switching equipment, circuit breakers, and transformers for
reducing voltages from high-voltage lines to those of lower voltage
(e.g., from 345 kV to 138 kV).
The electricity distribution system begins at the level of the
distribution substations. Here the high transmission voltages (69 kV
and above) are reduced to voltages used in the primary distribution
feeders (2.4 kV to 34.5 kV). Those substations may also contain circuit
breakers and switching equipment.
From the substation, the primary feeders are constructed in a
radial fashion, typically via overhead power poles and lines, and are
connected to major and minor primary lateral lines that supply the cus~
tomers. Near each customer is a transformer that reduces the distri-
bution voltage to 120/240 V for use in homes and business.
5. Coal Gasification Facility
a. Background
Because System 1 is a dual energy mode system, a parallel
system for supplying residences with pipeline gas is required in addi-
tion to the electricity production and distribution system. This system
is based on the production of synthetic natural gas (SNG) from coal.
IV-16
-------
The facility for producing SNG is assumed to be located near the coal
mine. Thus, coal is transported directly from the mine to the coal
storage facility for the SNG plant.
The facility we have chosen to analyze produces
3
7.8 million nm (275 million scf) of SNG per day. Such a plant would
consume 5.6 million tonnes (6.2 million tons) of subbituminous coal per
year, equivalent to about 1.25 times the output of the hypothetical
4.5 million tonne per year (5 million ton per year) mine.
Coal can be converted to SNG by a gasification process
followed by clean-up and methanation processes. Coal gasification has
been practiced on a commercial scale since the early part of the last
century. However, early coal gasification did not produce a high heat-
ing value SNG for long distance transmission in high-pressure pipe-
lines. To be transmitted economically, gas must have a high heating
value. The processes necessary to convert the product of the gasifier
to gas with a high heating value (shift conversion, acid gas removal,
and methanation) are currently being adapted from other industrial
processes.
There are two generations of coal gasification technol-
ogy. The first-generation processes are being operated on a commercial
scale. Perhaps the best current examples are the fixed-bed Lurgi coal
gasifiers at the South African Gas and Oil Co. (SASOL) plants and the
entrained-flow Koppers-Totzek gasifiers associated with the manufacture
of fertilizer (ammonia). Those processes have proved to be more expen-
sive and less efficient than desired. The developing second-generation
processes promise to improve upon the shortcomings of the first gener-
ation.
The Hygas process for the manufacture of SNG is an exam-
ple of a second generation coal gasification process followed by the
necessary upgrading processes to produce SNG. The Hygas process has a
potential economic advantage over the current, first-generation
IV-17
-------
gasification processes (Lurgi, Winkler, Koppers-Totzek). Hygas was not
selected for this study as necessarily the best technology, but as an
example of a second-generation process likely to become practical in the
next 10 to 20 years. Investment estimates and process flows have been
14
adapted from a study by C. F. Braun.
A block flow diagram of the SNG production plant is shown
in Figure IV-5. The principal processes are coal storage and prepara-
tion, coal gasification, shift conversion, acid gas removal, methana-
tion, and product drying. Other processes and facilities providing sup-
port are the cryogenic oxygen plant, steam and power production, the
plant water system, sulfur recovery, solids disposal, and water reclama-
tion (sour water stripping, ammonia recovery, and biological oxidation).
b. Process Description
The process flow is illustrated by the Hygas process flow
diagram, Figure IV-5, and the accompanying list (Table IV-3) of major
stream flow rates and compositions.
Coal is delivered from the mine on belt conveyors. After
sampling and magnetic removal of tramp iron, the coal is sent to a 30-
day storage pile. The pile serves as a surge damper to mitigate upsets
in the mine or delivery systems and as a method of averaging the feed
compositions. The coal is usually stacked and reclaimed in patterns
that mix or average the composition of coal. The mixing reduces process
operating problems that result from the rapidly changing characteristics
of feedstock.
Coal from the storage pile is sent to both the steam
boiler and the grinding mill for preparation of the process feed. Two
parallel coal preparation trains are used; each processes 50% of the
feed. Ground coal is moved in a water slurry from the mill to the high-
pressure slurry pump. Surge tanks and mixers minimize fluctuations in
the flow rate and allow adjustment of the content of the slurry water.
IV-18
-------
f
(-•
vO
RAW WATER
COAL
from mine
SNG
CO2 TO PLANT
INERT GAS AND
ATMOSPHERE
BIOLOGICAL.
SLUDGE
TO MINE
BY-PRODUCT
SULFUR
FLUE
GAS TO
ATMOSPHERE
FIGURE IV-5. BLOCK FLOW DIAGRAM OF A HYGAS SNG PLANT
-------
Table IV-3
MAIN PROCESS FLOWS FOR SNG FROM COAL VIA THE HYGAS PROCESS
Strewn Name and Number
Component
Quantity
12345 67 8 9
Coal Boiler Gasi- Water to Oxygen Steam Water to
From Feed Process fier Slurry to Gas- to Gas- Char Ash Char Ash
Mine Coal Feed Feed Tank ifier ifier Slurry Slurry
Dry coal
Water with
coal
Char & ash
V
H2
CO
co2
CH4
C2H6
103 kg/hr 560
103 kg/hr 160
103 kg/hr
103 g-mole/hr
103 g-mole/hr
103 g-mole/hr
10 g-mole/hr
10 g-mole/hr
103 g-mole/hr
Oxygen
Benzene
V
Solids
Temperature
Pressure
10 g-mole/hr
10 g-mole/hr
103 g-mole/hr
103 g-mole/hr
10 g-mole/hr
10 kg/hr
kPa
(psig)
47
13
515
145
515
570 425
28,000 8,200
50
8,200
3,500
290
(550)
8,820
(1,265)
300
(580)
8,820
(1,265)
(Continued)
IV-20
-------
Table IV-3 (Concluded)
Stream Name and Number
Component
Dry coal
Water with
coal
Char & Ash
H20
H2
CO
C°2
CH4
C2«6
NH3
H2S/COS
Oxygen
Benzene
V
Solids
Temperature
Pressure
Quantity
103 kg/hr
103 kg/hr
103 kg/hr
103 g-mole/hr
10 g-mole/hr
10 g-mole/hr
10 g-mole/hr
103 g-mole/hr
103 g-mole/hr
103 g-mole/hr
3
10 g-mole/hr
10 g-mole/hr
103 g-mole/hr
10 g-mole/hr
103 kg/hr
°C
(°F)
kPa
(psig)
10
Raw Gas
—
—
—
44,000
11,000
9,400
9,500
6,500
500
210
110/4.4
—
76
150
1.9
315
(600)
8,290
(1,200)
11
Quenched
Raw Gas
—
—
—
25,000
11,000
9,400
9,500
6,500
500
170
110/4.4
—
76
150
—
238
(460)
8,160
(1,185)
12 13 14 15
Conden- Cooled
sate Shifted Conden-
Slurry Slurry Gas sate
—
— — __ —
— — — —
19,000 590 35 20,000
5 16,000
3 4,700
8 2 14,000
4 — 6,500
500
39 1 — 170
11- — 100/2.2 10
_____
_
(Cfi+)64 160
1.9 —
238 260 38 332
(460) (500) (100) (630)
8,160 827 7,580 8,100
(1,185) (120) (1,110) (1,175)
16 17 18 18
Scrubbed Methanated Vent
Gas Gas SNG Gas
__
—
—
2 13 1 —
16,000 1,600 1,600
4,700 14 14
220 220 220
6,500 12,000 12,000 1
410
_
_
—
__
_
__
343 38 41
(650) (100) (105)
7,440 6,990 6,960
(1,080) (1,015) (1,010)
-------
Two 50% capacity coal feeding trains are used. Each
train uses three reciprocating pumps with one spare pump. The high-
pressure (9,470 kPa or 1,360 psig) coal slurry is preheated to 290°C
(550°F) by heat exchange with quenched raw gas and shift gas effluent.
The coal slurry is then injected into the fluidized top
bed of the gasifier (reactor), where the slurry water is flashed off at
315°C (600°F). The average residence time in the bed is 15 minutes
(all residence times were estimated by the Institute of Gas Technology).
Dried coal then falls to the entrained bed low-temperature (740 C or
1,360°F) reactor, where the residence time is about 10 seconds. This
low-temperature gasification-devolatilization-pyrolysis is partly
responsible for the high efficiency of the Hygas process. The less
reactive coal is gasified in the fluidized high-temperature (940 C or
1,720°F) reactor with an average residence time of 44 minutes. Ash
and char not reacting in the parts of the reactor described so far fall
into the very high temperature (1,000°C or 1,850°F) fluidized bed
steam-oxygen gasifier. Most of the remaining carbon is either gasified
by the steam carbon reaction (H20 + C—"-H- + CO) or burned with
oxygen (C + 0—»-CO_) to provide heat for the endothermic steam-
carbon reaction. Ash from the gasifier is quenched with water in a
spent char slurry tank. The char ash slurry is depressurized, cooled,
and sent to the solids disposal unit.
Two reactors would be needed in an SNG installation, each
handling 50% of the process flow. The reactor vessels would be huge,
even compared with large modern oil refinery equipment. Overall height
would be 67 m (220 ft) with a maximum diameter of 7.3 m (24 ft).15
The vessel walls would be from 13 to 18 cm (5 to 7 in.)
thick, and the gasifier would weigh about 1,800 tonnes (2,000 tons).
Such a large size and heavy weight precludes any significant construc-
tion in the shop and requires that it be built in the field, which is
more time-consuming.*°
IV-22
-------
The vessel would have dry-wall construction (no external
water jacket for cooling). The "low-alloy" (1% Cr and 0.5% Mo) shell
will be clad with a stainless steel (18% Cr, 8% Ni) liner and have re-
fractory lining for high-temperature zones.
Product gas from the gasifier goes through a cyclone to
remove carry-over coal dust, which is injected into the high-temperature
zone of the gasifier. Product gas at 315°C (600°F) is then directly
quenched with water to remove more entrained solids. Indirect heat ex-
change with the reactor feed slurry and boiler feed water reduces the
gas temperature to 240°C (460°F). At that temperature, the water
vapor concentration is 39 mole percent—at the proper level for shift
conversion. In the shift process, about one-half of the gas is bypassed
around the reactors. The shift is accomplished with two 50% capacity
shift trains. Each train has two reactors of equal size. The shifted
gas is combined with the bypassed stream and cooled in a series of steam
generators and boiler feed water heaters. Condensate formed by cooling
the shifted gas is rich in ammonia and is sent to the effluent treating
section. The shifted gas then goes to the acid gas removal plant.
The acid gas (H S and C0») removal process (such as
Allied Chemical's Selexol process or Lurgi's Rectisol process) separates
two streams from the methanation feed. One stream is rich in H_S, and
the other is principally C02. Separating the bulk of the C02 from
the H_s stream simplifies the sulfur recovery unit design. Because
details of the processes are proprietary, no process flow diagrams are
18
included; however, a brief description of typical applications fol-
lows. The process is dependent on physical absorption, with the di-
methyl ether of polyethylene glycol or cold methanol as a solvent.
H S and C02 are absorbed into the solvent at high pressure. By low-
ering the pressure of the "rich" solvent in successive stages, it is
possible to release fuel gas, CO., and H«S from the solvent. Final
stripping of the solvent can be done in a column with a heated feed.
After acid gas removal, the feed goes to the methanation
section and is heated to 340°C (650°F) before passing through a zinc
IV-23
-------
oxide sulfur-scavenger bed. This guard bed is designed to protect the
sulfur-sensitive methanation catalyst from minor amounts of feed sul-
fur. Because of the large exothermic heat of reaction of methanation,
the gas is reacted in three serial reaction stages. The product from
each stage is cooled in waste heat boilers. Four parallel sets of meth-
anation reactors (12 reactors in all) are required. To reduce the tem-
perature rise in the methanation reactors, a large product gas recycle
stream (about twice the flow rate of the net methanation product) is
used to dilute the methanation feed. The recycle compressor for this
diluent stream has a 340 C temperature at the inlet and requires
careful engineering design. Methanation product gas is dried in a tri-
ethylene glycol dehydration unit and sent into the product SNG pipeline.
Some significant process variations are currently under
development. One variation is a combined shift and methanation process,
which would be performed after the acid gas is removed. Chem Systems
and Ralph M. Parsons Co. (RMP) are both developing combination pro-
cesses. The RMP process offers the potential of more efficient heat
recovery from the methanation reaction. For this study, those combined
shift-methanation processes should be considered process variations un-
likely to significantly change the cost or efficiency of SNG production.
6. Gas Pipeline
To transport the SNG produced in the Powder River Basin to
markets in the Midwest requires a pipeline approximately 1,300 km (800
mi) long. New long-distance interstate pipelines are generally 81 or 91
cm (32 or 36 in.) in diameter, with corresponding capacities of 23 and
o
28 million nm (800 and 1,000 million scf) per day, respectively.
Those capacities represent the output of three or four 7.8 million nm3
(275 million scf) per day SNG plants. To supply a single large inter-
state pipeline, the SNG plants would deliver gas to smaller capacity
branch lines, which would feed the larger line. A branch line of 50 cm
(20 in.) in diameter could handle the output from a 7.8 million nm3
per day facility.
IV-24
-------
Because no major interstate pipelines currently connect the
Powder River Basin with the Midwest, new pipelines have to be built if
an SNG industry with the intent of supplying midwestern and eastern mar-
kets is to develop in the region. Likewise, a branch line connecting
each SNG facility to a larger interstate line is required.
We assume that the 1,300-km interstate pipeline is 81 cm in
o
diameter, has an approximate capacity of 23 million run per day, and
crosses one river and three roads per 160 km (100 mi) of length. Input
pressure at the coal gasification plant is 6,900 kPa (1000 psi). Pres-
sure is maintained along the pipeline by 11 compressor stations. Each
station has a 66.6 GJ/hr (24,800 horsepower) average centrifugal com-
pressor driven by a gas turbine. Fuel for the gas turbines is taken
from the SNG in the pipeline. A standby turbine-compressor is in each
station. In the past, reciprocating compressors have been used in place
of centrifugal compressors. Even though less efficient, the centrifugal
machines have become more popular because of greater simplicity and
greater reliability. Compressor stations are carefully instrumented,
highly automated, and frequently unmanned.
Typical pressure in the pipeline varies from 7,600 kPa (1100
psi) at the compressor discharge to 3,450 kPa (500 psi) at the compres-
sor suction. If the pipeline has sufficient strength, it can be used
for gas storage. Called line packing, this storage technique is em-
ployed by raising the average pressure of the pipeline. The normal ca-
pacity of the 1,300-km pipeline is about 5 days' production of a
3
7.8 million nm per day SNG plant. By line packing, the line storage
capacity can be increased 50 to 100%.
7- Gas Distribution
At the point where the pipeline passes nearest the city where
the gas is to be delivered, a city gate station transfers gas into the
local distribution system at 790 to 1,140 kPa (100 to 150 psig). Beyond
IV-25
-------
this point is a system of gas mains, valves, regulators, and meters that
controls and transmits the flow of gas to the ultimate users. As the
distribution network becomes more dense, the gas pressure is reduced,
finally reaching a delivery pressure of 103 to 104 kPa (0.25 to 0.35
psig) for residential customers.
8. Gas Furnace and Air Conditioner
The heating and cooling equipment employed in residences must
be sized to meet the peak heating and cooling loads expected throughout
the year. Those loads are dependent on both the temperature extremes
encountered in the geographic area under consideration and the thermal
properties of the residences. We assume, for the purpose of analysis,
that all residences supplied by the energy supply systems have the same
thermal characteristics. The characteristics are based on a study by
19
Westinghouse for the Electric Power Research Institute.
The residences are split-level, wood frame houses with
2 2
136 m (1,460 ft ) of interior floor area excluding the garage area
2 2
of 36 m (390 ft ). The houses are insulated to FHA minimum
standards; that is, insulation values of R-ll in the walls, R-19 in the
ceiling, and R-7 over the garage. Air infiltration rate is one complete
air change per hour. Based on those parameters, and other features of
the houses such as window area, the heat loss parameter is 360 kJ/hr-°C
(610 Btu/hr-°F). That means that 360 kJ/hr is lost from the house
for each °C difference between the interior and exterior tempera-
tures. If we assume that the interior temperature is maintained at
21 C (70°F) and calculate an average internal heat load (people,
lights, and appliances) of 4,870 kJ/hr (4,620 Btu/hr), then the heat
load, Q, as a function of exterior temperature T (°C or °F), is
found to be
Q = 39,900 - 360T kJ/hr
(Q = 37,800 - 610T Btu/hr).
IV-26
-------
Thus, the heat load for an extremely low temperature encoun-
tered during the winter heating season in the Omaha-Des Moines-Kansas
City region — say, -29°C (-20°F) — would be 53 MJ/hr
(50,000 Btu/hr). The gas furnace chosen in the Westinghouse study for
use in the residences described above delivers 69.6 MJ/hr (66,000
Btu/hr) with a fuel input of 87.0 MJ/hr (82,500 Btu/hr), and thus has
ample capacity to meet the heat load.
The procedure for choosing the appropriate air conditioner is
somewhat more complicated because the cooling load depends on factors
such as solar heat input and relative humidity, as well as external tem-
perature. Air conditioner size is based on a design point of tempera-
ture and humidity not likely to be exceeded more than 2.5% of the time
during the summer months (June through September). The design condi-
tions for the region under consideration (using Omaha data as average
for the three cities) are a dry bulb temperature of about 34°C
(94°F) and a wet bulb temperature of about 26°C (78°F). The wet
bulb temperature is a measure of the relative humidity, which, for the
two temperatures just mentioned, is 49%. The cooling load calculated
for those conditions is 25.0 MJ/hr (23,700 Btu/hr), assuming the inte-
rior temperature is to be maintained at 26°C (78°F). (See Chapter
VIII.)
An air conditioner that can meet that cooling load is the Wes-
tinghouse SL030C/EC030, with a cooling capacity of 29.9 MJ/hr (28,300
Btu/hr), rated at 35°C (95°F) exterior dry bulb temperature and
26°C (78°F) dry bulb, 19°C (67°F) wet bulb interior return air
temperature. Because cooling capacity increases with decreasing exte-
rior temperature, capacity is clearly ample at the design condition for
this choice of air conditioner.
The cooling capacity of the Westinghouse SL030C/EC030 air con-
ditioner is shown in Figure IV-6 as a function of temperature, along
with the cooling load calculated for average summer afternoon humidity
and solar heat input. (The calculation of the cooling load is discussed
in Chapter VIII.)
IV-27
-------
M
to
oo
30
Q
Z
LU
Q
O 20
a.
a.
c/j
O
O
O
10
1
75
80
EXTERNAL TEMPERATURE -°F
85 90 95
T
100
105
110° f
_L
r
j_
25
30,000
20,000 2
CO
10,000
30 35
EXTERNAL TEMPERATURE -°C
40
45° C
FIGURE IV-6. COOLING CAPACITY OF THE WESTINGHOUSE SL030C/EC030 AIR CONDITIONER
-------
B. System 2
A block flow diagram of System 2 is shown in Figure IV-7- The com-
ponents of the system are described below.
1. Coal Mine
See Section IV-A for the complete description of a surface
coal mine located in the Powder River Basin.
2. Coal Gasification Facility
See Section IV-A for the complete description of the Hygas
coal gasification process.
3. Gas Pipeline
See Section IV-A for the complete description of the inter-
state natural gas pipeline.
4. Gas Distribution
The distribution of SNG or natural gas to a 26-MW fuel-cell
power plant (described in the following section) is considerably dif-
ferent from the distribution to individual residences described in
o
Section IV-A6. Whereas a residence may consume 140 to 280 nm (500 to
1,000 scf) of gas per day during the winter season (with a gas-fired
2
furnace), a 26-MW power plant consumes on the average 53,800 nm (1.9
x 10 scf) per day. That amount represents one-quarter of 1% of the
capacity of the 81 cm (32 in.) interstate pipeline.
IV-29
-------
COAL MINE
COAL
GASIFICATION
PLANT
GAS PIPELINE
GAS
DISTRIBUTION
26-MW
FUEL CELL
POWER PLANT
ELECTRICITY
DISTRIBUTION
HEAT PUMP
FIGURE IV-7. BLOCK FLOW DIAGRAM OF SYSTEM 2
IV-30
-------
Fuel-cell power plants located within a city would be served
by individual lines originating at the city gate. Such a system would
o
be similar to that used for supplying large (greater than 71 nm or
2,500 scf per hour) commercial and industrial customers.
5. 26-MW Fuel-Cell Power Plant
a. Background
In this system, we assume that dispersed fuel-cell power
plants provide intermediate-load electric power to residential custom-
ers. To represent the most advanced technology likely to be available
for this application in the late 1980s or early 1990s, we assume that
such power plants use molten carbonate fuel cells. This technology was
chosen because it represents advancement in both cost and efficiency
over first-generation (phosphoric acid) fuel-cell technology, and be-
cause it more effectively illustrates the advantages of the concept of
fuel cells.
The choice of this technology is justified by the follow-
ing considerations. First, although phosphoric acid technology will be
used to prove the effectiveness of fuel cells in utility applications,
and to achieve initial market penetration, because of inherent
limitations in the technology, further advances in costs and performance
will probably not be dramatic. Second, achieving widespread utilization
of fuel-cell power plants will require technology considerably more
advanced than that represented by phosphoric acid. Because present
molten carbonate technology is more susceptible to dramatic improvements
than phosphoric acid technology it is the only realistic candidate for
this study.
The fuel-cell power plants described in this section are
designed to achieve the DOE-EPRI-UTC heat rate goal of 7,910 kJ/kWh
(7,500 Btu/kWh). They are also designed to be water-conservative—that
is, no liquid water is delivered to the site of the power plant, nor
IV-31
-------
discharged from it. The only outside material supplied is the SNG fuel,
other than occasional maintenance materials, periodic replacement of the
zinc oxide guard bed, and, after 40,000 hours of operation, refurbish-
ment of the fuel-cell stack.
The description of the fuel-cell power plant in this
section is much more detailed than the description of other system com-
ponents, because other system components represent either well-known,
accepted technology (e.g., pipelines) or more advanced technology (e.g.,
coal gasification), which have been analyzed in great detail in other
studies. Fuel-cell power plants of the kind considered here represent
truly new technology for which little detailed analysis has been carried
out. Thus, we have specifically analyzed fuel cell design and power
plant configuration to effectively assess the environmental and economic
feasibility of using fuel cells in the applications specified in Chapter
III.
b. System Description
A flow plan for the 26-MW molten carbonate fuel cell
power plant was developed, and is shown in Figure IV-8. This integrated
system was designed to achieve the target heat rate and to be water con-
servative. Molten carbonate fuel-cell technology and systems concepts
are at a relatively early stage of development, however, and alternative
20
approaches could be used to satisfy the imposed target goals . The
nonoptimized design used in this analysis can be considered as represen-
tative. Full optimization was beyond the scope of this study.
The system is best understood by following the SNG
stream. The SNG supply (Stream 0) contains small amounts of mercaptans,
most of which can be removed by a zinc oxide adsorption bed. The desul-
furized SNG (Stream 1) and recycled water (Stream 5) are preheated in a
series of heat exchangers E-l, E-2, and E-3. They enter the reformer
superheated relative to the reformer outlet temperature.
IV-32
-------
f
AIR
FIGURE IV-8. BLOCK FLOW DIAGRAM FOR 26MW FUEL CELL POWER PLANT (SNG FUEL)
-------
Reformed SNG (Stream 9) must be cooled before entering
the fuel cell anode compartment (Stream 10). The heat is used to pre-
heat the reformer feeds in E-2. In the fuel cell, H~ and CO are oxi-
dized to H_0 and C0_, and a stoichiometric amount of C0« is added
from the electrolyte. The unused anode fuel (Stream 11) is burned with
preheated air (Streams 12, 13, and 14) to supply heat for the reformer
(Stream 15). The hot burner gas leaves the reformer (Stream 16) and is
cooled down to condense enough product HO to maintain a net water
balance in the system (Stream 5). Heat from Streams 16, 17, and 18 pre-
heats the reformer feeds in E-3, the burner air feed in E-4, and again
the reformer feeds in E-l. The final cooldown and condensation of
Stream 19 is accomplished by air in exchanger E-7. A knockout drum sep-
arates Stream 20 into the water recycle (Stream 5) and saturated gas
(Stream 21).
Part of the cooling air (Stream 25) combines with Stream
21 to become make-up feed for the cathode (Stream 26). Stream 26 is
preheated in E-6 and Stream 27 is then blended with a cathode recycle
(Stream 31) to make the cathode feed (Stream 28). In the cathode, 1/2
0 and CO- are stoichimetrically reacted into the electrolyte. The
cathode exhaust (Stream 29) is used to preheat burner air (Stream 12) in
E-5, and is then partially recycled via Stream 31 to the cathode. The
remaining cathode exhaust (Stream 32) is used to preheat the cathode
make-up feed (Stream 26) in E-6.
Many modifications of this design, both major and minor,
could also meet the design goals. These designs, however, would also
require a fairly elaborate system of thermal integration.
c. System Design Basis
Several points in the design of this power plant deserve
further discussion to clarify the assumptions and constraints of the
design:
IV-34
-------
Sulfur in the SNG fuel, in the form of odorant mercaptans, can
gradually poison both the reformer catalyst and the fuel cell
electrodes. The zinc oxide bed must lower the sulfur content
below 1 ppm.
SNG reforming is highly endothermic, so the reactor is limited
by heat transfer rather than catalyst activity. The reformer
requires high temperature heat. The reformed SNG composition
can be estimated, based on the equilibrium between Ctfy and
the shift reactants, CO, C02, H2, and
o Low methane slip (unconverted CIfy) in the reformer is
desirable. CH^ is not active in the anode reaction and
sequesters potential anode fuel (H2 and CO).
o Heat for the reformer is best supplied by burning unused anode
fuel. Anode waste is preheated and the most active components
have already contributed to electrochemical energy genera-
tion. Direct burning of SNG would need additional preheat
heat exchangers and would not contribute to the DC output.
Preheating both the reformer feed (Streams 4 and 8) and the
burner air (Stream 14) reduces the fuel requirement for the
reformer.
o C02 is a major reactant at the cathode and dramatically af-
fects cell voltage. C02 is conveniently supplied to the
cathode by the burned anode exhaust.
o The cathode inlet must be preheated. The air for exchanger
E-7 (Streams 23 and 24) has been partially preheated.
o The cathode recycle (Streams 29, 30, 31, and 32) recycles heat
and keeps the average concentration of C02 and 02 in the
cathode compartment at higher levels than a single pass
cathode.
o The fuel-cell power density is increased by lowering the out-
put voltage and by reducing the consumption of reactants at
both the anode and the cathode.
o Water conservation is somewhat easier to achieve by conden-
sation from the burned anode effluent rather than from the
cathode effluent. H20 is more concentrated in Stream 19 so
that less final cooling is necessary. Also, 1^0 acts as a
diluent at the cathode. Removing 1^0 from the cathode feed
improves cathode activity.
SNG refonnate composition was estimated using procedures
developed by Imperial Chemicals Industries (see Section IV-E) . System
heat balances were computed using component enthalpy data published by
IV-35
-------
21
the Girdler Corportion x. The calculation procedure is outlined in
Appendix A.
The effect of reactant composition and utilization, oper-
ating pressure, and cell design parameters on molten carbonate fuel cell
performance were estimated using an approximate analytical procedure
developed at Exxon Research and Engineering by H.H. Horowitz. This pro-
cedure is detailed in Appendix B.
d. System Operating Characteristics
Several iterative calculation cycles were carried out to
define a set of system operating parameters that would meet the perfor-
mance targets. The final system material and heat balance is summarized
in Table IV-4. Corresponding fuel-cell performance was estimated as
o
0.8 V/cell at an average current density of 120 mA/cm . Combined H~
and CO fuel utilization at the anode was 77%. Near atmospheric pressure
operation was selected, based on predictions showing small effects on
cell performance for elevated pressure operation. High-pressure opera-
tion is feasible for this system, using a coupled turbocompressor/
expander set in the air loop. However, cost savings resulting from
smaller line sizes at higher pressure were assumed to be negligible.
e. Conceptual System Design
A conceptual design was prepared for the System 2 power
plant components and assembled layout. The equipment list, showing ca-
pacity, size, and materials of each component, is given in Table IV-5.
Modular configurations are envisioned, with components sized to permit
factory assembly and subsequent shipment in individual trailers. That
requirement determines the capacities of the fuel-cell trailer and the
reformer/heat exchange package, which are the largest and heaviest com-
ponents in the system. The modular sizes are shown in columns 4 and 5
of Table IV-5.
IV-36
-------
Table IV-4
PROCESS FLOW STREAMS FOR 26-MW FUEL-CELL POWER PLANT (SNG)
Temperature
Stream
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
°C
15.6
15.6
471
593
760
74.4
471
593
760
702
593
704
15.6
593
704
1,160
849
774
759
287
74.4
74.4
15.6
65.6
65.6
65.6
70.2
322
557
704
672
672
672
279
(°p)
(60)
(60)
(879)
(1,100)
(1,400)
(166)
(879)
(1,100)
(1,400)
(1,295)
(1,100
(1.300)
(60)
(1,100)
(1,300)
(2,120)
(1,560)
(1,425)
(1,399)
(548)
(166)
(166)
(60)
(150)
(150)
(150)
(159)
(690)
(1,035)
(1,300)
(1,241)
(1,241)
(1,241)
(534)
Enthalpy
(GJ/hr)
-12.92
-12.92
-8.05
-6.34
-3.78
-219.9
-174.2
-170.3
-164.8
-131.0
-136.6
-496.9
3.68
14.6
16.8
-480.1
-517.6
-525.7
-527.9
-578.6
-632.6
-412.8
-220.8
274.8
260.0
14.8
-397.9
-360.6
-611.0
-346.9
-357.8
-250.5
-107.3
-144.7
H2 H20(g) CH4
27.74 0
27.74 0
27.74 0
27.74 0
27.74 0
__
817
817
817
714.1 521.
714.1 521.
164.2 1,071
6.
6.
6.
1,259.
1,259.
1,259.
1,259.
1,259.
442.
442.
377.
377.
356.
.3 204.0
.3 204.0
.3 204.0
.3 204.0
.3 204.0
—
—
—
—
2 9.18
2 9.18
9.18
30
30
30
9
9
9
9
9 — ~
7
7
1
1
9
Flow Rate
CO C02
0.24 3.
0.24 3.
0.24 3.
0.24 3.
0.24 3.
„
..
„
._
94.4 104.
94.4 104.
21.7 799.
-_
—
„
830.
830.
830.
830.
830.
830.
830.
„
„
„
, 10 g-moles/hr
74
74
74
74
74
7
7
9
8
8
8
8
8
8
8
20.27
463.
463.
1,543.
1,543.
1,543.
0
0
2
2
2
1,080.2
463.0
-- 463.
0
830.
830.
1,316.
694.
694.
486.
8
8
.6
,3
,3
,0
208.3
208,
,3
°2
—
—
—
—
—
—
—
—
—
—
—
—
.121.5
121.5
121.5
11.05
11.05
11.05
11.05
11.05
11.05
11.05
7,323.2
7,323.2
6,917.3
393.6
404.7
404.7
622.5
311.3
311.3
217.9
93.4
93.4
N2 H20(l) Total stream
—
—
—
—
—
—
—
—
—
—
—
—
489.8
489.8
489.8
489.8
489.8
489.8
489.8
489.8
489.8
489.8
29,293
29,293
27,718
1,574.4
2,064.2
2,064.2
6,880.7
6,880.7
6,880.7
4,816.5
2,064.2
2,064.2
236 . 1
236.1
236.1
236.1
236.1
817 817
817
817
817
1,443.5
1,443.5
2,066.0
617.7
617.7
617.7
2,591.6
2,591.6
2,591.6
2,591.6
2,591.6
817 2,591.6
1,774.4
37,148
37 , 148
35,004
1,988.3
3,762.5
3,762.5
10,363
9,429.4
9,429.4
6,600.6
2,828.8
2,828.8
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
IlV^-37
-------
Table IV-5
EQUIPMENT LIST FOR 26-MW FUEL-CELL POWER PLANT (SNG)
OJ
00
Unit
Fuel Cell Trailer
E-l
E-2
E-3
E-4
E-5
E-6
E-7
Reformer Section
Reformer Catalyst
Burner
ZnO Bed
Knockout Drum
Condensate Pump
B-l
B-2
B-3
B-4
B-5
Capacity
26 . 7 MH D C
50.5 GJ/hr
5.6 GJ/hr
8.1 GJ/hr
2.2 GJ/hr
10.9 GJ/hr
37.2 GJ/hr
54.0 GJ/hr
37.6 GJ/hr
714,140 g-mole/hr HZ
2,005,700 g-mole/hr
1.22 kg/day S
817,200 g-mole/hr HjQ
4.1 liter/sec.
229 m3/min
13,800 m3/min
2,460 m3/min
662 m3/min
741 m3/min
27
0.
0.
0.
0.
0.
1.
1.
0.
20
8( V
^X
99 (x
19 (x
38 (x
12 (x
18 (x
44 (x
93 (x
86 (x
.2m3
Size
103 m2)
103 m2)
103 m2)
103 m2)
103 m2)
103 m2)
103 H,2)
103 m2)
103 m2)
Modular Size
0
0
0
0
0
0
0
0
* 0
• *to
.22
.046
.096
.031
.045
.36
.48
.21
, ln3 Jl)
^x 1U m J
(x 103 m2)
(x 103 m2)
(x 103 m2)
(x 103 m2)
(x 103 m2)
(x 103 m2)
(x 103 m2)
(x 103 m2)
See
102
122
6.7
See
Dimensions
Figure IV-10
cm dia x 3.7 m long
cm dia x 7.9 m long
x 9.1 x 4.6 m high
Figure IV-10
504 m3
169 GJ/hr
1.
7.
8 m3/h
25m3
r
42.2
0
1
3
.46
.18
1
57
,450
616
166
185
GJ/hr
m3
m3
.0 liter/sec
.3 m /min
m /min
m /min
m /min
m /min
See
Figure IV-10
0.61 m dia x 2.1 m high
1.3
0.61
m dia x 2.1 m long
x 0.61 x 1.3 m long
0.91 x 0.91 x 0.91 m
Part
1.5
1.3
1.3
of E-7
x 1.5 x 1.8 m
x 1.3 x 1.3 m
x 1.3 x 1.3 m
Materials
UTC Advanced
304 SS
309 SS
304 SS
304 SS
304 SS
304 SS
C.S.
HK 40
Design
Girdler G-56A
C.S. & Refractory
Girdler G-72
Galvanized
C.S.
C.S.
C.S.
304 SS
C.S.
C.S.
-------
The limiting size for normal transport by truck is 3.8 x
3.8 x 12 m (12.5 x 12.5 x 40 ft), with a maximum weight of 31,800 kg (35
tons). Railroad flat cars can carry units up to 3.8 x 3.8 x 15 m (12.5
x 12.5 x 50 ft) in size and 63,600 kg (70 tons) in weight. Based on
weight estimates, the proposed modular packages are shippable by rail,
but smaller modules could be designed if shipping by truck were consid-
ered necessary. The overall size and cost of the plant would be some-
what greater if smaller modules were used.
Fuel-Cell Trailer Design — The molten carbonate fuel-cell
stacks are mounted in shippable trailer modules. The design of the
fuel-cell section was based on the following parameters:
o Gross DC power output - 26.7 MW
o Cell current density - 120 mA/cm2
o Cell voltage - 0.800 V
o Total active area - 27,800 m2 (299 x 103 ft2)
o Cell active area - 0.85 m2 (9.16 ft2)
o Number of cells per stack - 510
o Total number of stacks - 64
o Stack voltage - 408 V (DC)
o Number of stacks per trailer - 8
o Trailer voltage - 816 V (DC).
The configuration of the fuel-cell trailer is shown in Figure
IV-9. Eight stacks, 1.2 x 2.4 x 1.2 m (4 x 8 x 4 ft) in size, are
mounted on a steel base structure with a steel framework to support the
upper row of four stacks. A possible gas manifolding arrangement is
indicated. Each row of stacks is connected electrically in parallel.
The two rows are connected in series, to give a trailer voltage of
816 volts DC. Higher voltages are possible by connecting more stacks in
series. The stacks are thermally insulated and the trailer, which
IV-39
-------
ANODE
FUEL IN
3.0m
ANODE
EXHAUST
OUT
'
FUEL MANIFOLDING
SIDE VIEW
OXIDANT MANIFOLDING
8 FUEL CELL STACKS
EACH 1.2 x 1.2 x 2.4m
3.7m
CATHODE
•-INLET
FLOW
CATHODE
-^OUTLET
FLOW
CATHODE
*. INLET
FLOW
FIGURE IV-9. FUEL CELL TRAILER LAYOUT
IV-40
-------
weighs about 44,500 kg (49 tons), is enclosed with corrugated panels.
The power plant requires eight fuel-cell trailers. Fuel to each pair of
trailers is supplied by one reformer package.
Reformer/Heat Exchanger Module Design — The reformer and heat
exchangers for the power plant were sized, based on discussion with
Exxon design engineers and information supplied by Girdler^. The
reformer was sized for 9.4 cm (3.7 in.) i.d. tubes at about 152 kPa
(1.5 atm) total pressure, using a space velocity of 1,600 m
o
gas/hr-m catalyst, based on the total feed rate of SNG and steam.
Small diameter tubes were chosen to minimize both the volume of catalyst
and total area of the tube. The reformer tubes are made of HK-40, a
high-alloy steel required by the temperatures and H partial
pressure. The reformer catalyst is Girdler G-56a, a standard methane
reforming catalyst (nickel supported on alumina).
Heat exchanger area estimates were based on overall heat
n
transfer coefficients, which ranged between 100 and 200 kJ/m - C-hr
(5 and 10 Btu/ft -°F-hr). The lower coefficient applies to heat
transfer by convective mechanisms between two air streams. That coef-
ficient is increased somewhat by the presence of H_ in one or both
streams, or by other heat transfer processes, such as radiation at high
temperatures, liquid convection, boiling or condensation. The heat
transfer coefficients were applied to 5 cm (2 in.) i.d. tubes. The
material specified for exchangers E-l, E-3, E-5, and E-6 is 304 stain-
less steel (SS) and 309-SS for exchanger E-2. The other heat exchange
tubes can be fabricated from normal carbon steel.
Pressure drops in the various units are extremely small —
less than 6.9 kPa (1 psi) in the reformer and a few kPa (less than 1
psi) in the heat exchangers. Thus, the entire system operates at near
atmospheric pressure. Pumps and blowers were specified by their flow
rates and pressure drops.
IV-41
-------
The zinc oxide sulfur trap was sized using data on Girdler
G-72, which can reduce gaseous sulfur to 0.2 ppm S and has a maximum
3 3
sulfur loading of 250 kg/m (15 Ib/ft ). The ZnO is non-regenerable
and must be periodically discarded. However, because the input level of
o
mercaptans in SNG is expected to be low—10 kg/million m (0.6
•i
Ib/million ft ) of gas—the yearly replacement of the zinc oxide bed
is a minor consideration.
This study's conceptual design of the reformer/heat exchanger
package is a unit that contains a burner, reformer section, and several
heat exchangers. The configuration is shown in Figure IV-10. The
burner is located at the base of a cylindrical furnace. The shell is
assumed to be fabricated from 0.64 cm (0.25 in.) thick carbon steel with
15.2 cm (6 in.) thick castable refractory inner lining. The reforming
section is positioned directly above the burner and consists of a series
of 10 cm (4 in.) o.d. tubes cast in HK 40 alloy steel (25 Cr-20 Ni).
The reformer tubes will most likely be positioned vertically to avoid
set- tling of the catalyst pellets, which would result in bypassing of
steam and SNG and incomplete conversion to H«.
Heat exchangers are of canal-type recuperator design. They
are located in the flue gas flow path above the reforming section. Heat
exchanger E-2 is mounted vertically on the side of the cylinder. Ex-
haust gas flows from the top of the unit and is ducted to the air fin
condenser (E-7).
This configuration is proposed as a compact package that can
be factory assembled and that would have relatively short sections of
ducting between components. This particular design was used as a basis
for estimating power plant size and investment costs. The package may
prove impractical, however, if experience shows that the reformer
section requires periodic maintenance. In that case, it would be better
to build the reformer as a separate unit, and have the heat exchangers
mounted on a separate platform. The optimized configuration can only be
determined after detailed designs and cost information are known.
IV-42
-------
FLUE GAS TO
CONDENSER E-7
BOILER/
SUPERHEATER
E-
WATER FROM
5 )- FEEDWATER
PUMP P-1
TO FUEL
10 y+ CELL
ANODE
4.6m
FIGURE IV-10. REFORMER/HEAT EXCHANGER PACKAGE
IV-43
-------
Table IV-5 lists the capacity, size, and materials of con-
struction for the various components in the package. The weight of the
package is approximately 45,500 kg (50 tons). Four reformer package
units are required for the complete power plant.
Equipment Module — The equipment module shown in Figure IV-11
contains the smaller components in the system. All components in an
individual equipment module were sized to service one reformer and two
fuel-cell trailers. The components are mounted on a steel-welded I-beam
base with space above used for fluid piping and ducting. The components
in the equipment module include: heat exchangers E-5 and E-6; the ZnO
tower; the knockout drum; blowers B-l, B-3, B-4, and B-5; and condensate
pump P-l. The overall dimensions of the equipment module are approxi-
mate.
Other configurations are certainly possible. This design uses
shell and tube heat exchangers for E-5 and E-6. However, because the
system pressures are very low, other types of heat exchangers could be
used. For example, either canal type or plate fin recuperator ex-
changers may be selected. That would result in somewhat different
module dimensions.
The power plant is designed to be water-conservative. How-
ever, make-up water may be required during periods of operation at
excessive ambient temperature. In this case, a small ion exchange bed
system could be provided in the equipment module to provide the clean
water necessary for reformer operation. This deionized water production
facility was omitted from the conceptual power plant design.
Power Plant Layout—A possible layout of one subsystem of
modules is shown in Figure IV-12. Four subsystems make up the total
plant. The overall plant size is estimated to be 39 by 55 m (128 by 180
ft). Considerable space is required for the air fin condenser E-7,
which transfers waste heat to the atmosphere. These units contain fans
(B-2) that pull atmospheric air (Stream 23) pressurized by B-5 and fed
back into the cathode inlet mixture (Stream 26).
IV-44
-------
FLUE GAS FROM £-7
f
FROM CATHODE
TO E-4
OVERALL DIMENSIONS: 3.7m WIDE
12.2m LONG
3.0m HIGH
FIGURE IV-11. EQUIPMENT MODULE LAYOUT
-------
19.5m
f-
CT*
EQUIPMENT MODULE
AIR IN
(j
/
V
\-
3)
\
FUEL
CELL
TRAILER
^
I
FUEL
CELL
TRAILER
C
POWER
CONDITIONER
27.4m-
OVERALL PLANT SIZE: 55m x 39m
FIGURE IV-12. POWER PLANT LAYOUT
-------
System piping was sized assuming an average gas velocity of
20.3 m/sec (4,000 ft/min). Piping specifications are listed in Table
IV-6. Insulation is required on air piping to prevent heat loss.
Table IV-6
PIPING SPECIFICATIONS
Stream No. Diameter, cm (in.) Length, m (ft) Material
5 5.1 (2) 22.3 (73) 302
10 66.0 (26) 18.3 (60) 304 SS
11 86.4 (34) 18.3 (60) 304 SS
13 43.2 (17) 38.4 (126) 304 SS
19 71.1 (28) 10.4 (34) Galvanized
20 45.7 (18) 18.3 (60) Galvanized
28 173. (68) 19.8 (65) 304 SS
29 198. (78) 9.2 (30) 304 SS
aSS = Stainless Steel.
6. Distribution of Electricity
Because fuel-cell power plants are intended for use as dis-
persed power sources, the arrangement for transmitting electricity to
the ultimate users is different than that described in Section IV-A for
remotely sited plants.
Most likely, electricity generated by dispersed fuel-cell
power plants will enter the utility grid at the substation level.
Therefore, no high voltage transmission will be associated with the
facilities. In essence, the fuel-cell power plant will serve as its own
substation, and will include transformers to step up the fuel-cell op-
erating voltage to that used in the local distribution system (13.8 kV).
Additionally, the appropriate switching apparatus and circuit breakers
will be provided for integration with the utility grid and isolation of
the fuel-cell equipment in case of an emergency.
IV-47
-------
7. Heat Pump
The residences supplied with electricity by the 26-MW fuel-
cell plant are heated and cooled by heat pumps. We assume that the heat
pumps are of advanced design, displaying performance characteristics
discussed below.
A heat pump is a machine that transfers heat from a lower tem-
perature reservoir to a higher temperature reservoir. Because it is a
heat transfer rather than a heat generating device, its "efficiency,"
measured as the heat supplied to the higher temperature reservoir di-
vided by the work required of the device, can be greater than unity.
Therefore, heat pumps can provide more space heating per unit of con-
sumed energy than combustion devices or electric resistance heating.
The effectiveness of heat pumps is measured by a parameter called the
coefficient of performance (COP), which is equal to the ratio of the
heat supplied to the electricity consumed. Modern heat pumps have COPs
ranging from 1.5 to 3 (depending on the exterior temperature, size of
the unit, and so on). The equivalent parameter for electric resistance
heat is 1.0, while for combustion furnaces it would be 0.5 to 0.7. A
particular advantage of heat pumps is that the same device that provides
space heating in the winter can provide air conditioning in the summer
by simply reversing the refrigerant cycle.
In the 1950s and 60s, when heat pumps first began to have a
sizable market, their reputation suffered because many early models had
low reliability and high maintenance costs. By and large, however, the
defects that caused problems in the early models have been corrected,
and models currently available are reliable and economical. As a result
of aggressive advertising by electric utilities to improve the heat
pump's image, their sales have grown at a dramatic rate in the past
several years. They are especially popular in areas where natural gas
is not available or has been curtailed, and in which electric resistance
heating and home heating oil are expensive alternatives.
IV-48
-------
Problems remain, however. Typically, the heat pumps are de-
signed to function in moderate climates where the air conditioning load
is larger than the space heating load. They are optimized for air con-
ditioning, and do not perform nearly as well in cold northern climates
where heating loads far exceed cooling loads. To achieve significant
market penetration in the North, they will have to be optimized for
northern seasonal conditions.
Under contract to the Electric Power Research Institute,
Westinghouse Electric Corportion carried out a set of optimization
studies to determine the changes in the design of present-day heat pumps
19
that would make them suitable for northern climates . A computer
program was constructed that modeled the performance and costs of major
heat pump components, including the compressor, evaporator coil, con-
denser coil, and air blower, with the sizes of these components vari-
able. A set of ten hypothetical heat pump designs were examined. These
heat pump designs resulted from a program that minimized the annual own-
ership costs (amortization, electricity cost, maintenance, taxes, and
insurance) for a particular set of conditions that included external
temperature, price of electricity, and changes in heat pump configura-
tions and use of advanced components (such as the compressor). The ten
designs were then subjected to a seasonal performance simulation for an
actual northern location (Albany, New York) using current electric
rates, to determine actual ownership costs, seasonal COP, and so on.
Of the ten designs proposed, several were comparable in terms
of performance and cost. However, one in particular appeared to be most
favorable for purposes of this study. That design (referred to as Heat
Pump No. 2 in the Westinghouse report) is changed with respect to cur-
rent models mainly in that the compressor is located indoors rather than
outdoors (and, of course, its components are sized for optimum heating
performance). That change greatly reduces the heat loss from the com-
pressor and allows heat that is lost to be recovered as part of the heat
supply. The layout of the components of that advanced heat pump system
is shown in Figure IV-13.
IV-49
-------
f
Compressor
Accumulator/Heat Exchanger
Outdoor Coil
Indoor Coil
Four-Way Valve
Heating Cap. & Check/ A P Valve
Cooling Cap. & Check/A P Valve
Outdoor Fan/Motor
Indoor Blower/Motor
10. Defrost Heater
11. Supplementary Heaters
1Z Oil Return Metered Line/Orifice/Etc.
13. Supply Duct
14. Return Duct
Source: Reference 19
FIGURE IV-13. RECOMMENDED HEAT PUMP CONFIGURATION FOR SPLIT SYSTEM AIR-TO-AIR UNITS
-------
Because heat pumps provide both winter heating and summer
cooling, sizing of the unit to meet a particular residence's heating and
cooling loads most economically is more complicated than for the gas
furnace/air conditioner case. The situation is simplified somewhat
because the heat pump need not be designed to meet the maximum heat
load; supplementary electric resistance heating can be used when re-
quired.
The customary practice in sizing heat pumps is to choose a
unit that can meet the maximum summer cooling load. In northern
climates, this size is generally too small to meet the maximum winter
heating load; enough capacity for electric resistance heating must be
provided to meet the load on the coldest winter day. In a region with a
very small cooling load, the unit should be somewhat oversized to mini-
mize the amount of expensive resistance heat required in the winter.
In Section IV-A, the design summer cooling load was given as
25.0 MJ/hr (23,700 Btu/hr) for 34°C (94°F) dry bulb, 26°C (78°F)
wet bulb temperatures. The cooling capacity of the optimized heat pump
(Heat Pump No. 2) analyzed in the Westinghouse study nearly matches that
load. It has a rated cooling capacity of 26.0 MJ/hr (24,600 Btu/hr) at
35°C (95°F) exterior temperature, and 26°C (78°F) dry bulb and
19 C (67 F) wet bulb interior temperatures. The heating and cooling
capacities of the heat pump as a function of temperature are shown in
Figure IV-14. The heating and cooling capacities reflect an assumed
21 C (70 F) indoor temperature for heating, and 26°C dry bulb and
19 C wet bulb indoor temperatures for cooling. Superimposed on those
curves are the heating and cooling loads for the residences described in
Section IV-A. The cooling load curve assumes average summer afternoon
conditions of humidity and insolation (see Chapter VIII).
As can be seen from Figure IV-14, the heating balance point
(the point at which heating capacity equals the heat load) is -1°C
(30 F). Below that temperature, supplementary electric resistance
heating must be used. Resistance heating would be provided in units of
IV-51
-------
70
Z 30
5
UJ
O
QC
O w 20
Z
O
10
EXTERNAL TEMPERATURE - ° F
80 90 100 110
i
"20 25 30 35 40
EXTERNAL TEMPERATURE - °C
30,000
20,000
3
+-<
CO
10,000
45
-20
-10
EXTERNAL TEMPERATURE - °F
10 20 30
40
60
-25
SUPPLEMENTARY ELECTRIC
RESISTANCE HEATING
50
-20
-15 -10 -5 0
EXTERNAL TEMPERATURE - °C
10
60
60,000
50,000
40,000
30,000 f
20,000
10,000
15
FIGURE IV-14. HEATING AND COOLING CAPACITY OF 26.0 MJ/hr (24,600 Btu/hr)
HEAT PUMP
IV-52
-------
4.7 kW (16,000 Btu/hr) each. When the temperature falls below -1°C,
the units are cycled on and off as required to maintain the interior
temperature at 21°C. Three units of 4.7-kW capacity are sufficient to
maintain that temperature down to an exterior temperature of -29 C
(-20°F). The additional capacity added by the resistance heaters is
shown by dashed lines in Figure IV-14.
C. System 3
A block flow diagram of System 3 is shown in Figure IV-15. The
components of this system are described below.
1. Coal Mine
See Section IV-A for the complete description of a surface
coal mine located in the Powder River Basin.
2. Coal Liquefaction Plant
a. Background
Coal may be converted to liquid fuels by several pro-
cesses currently under development. H-Coal is the liquefaction process
selected for review in this study because it can be modified to produce
only distillate fuels and it is one of the processes most likely to be
commercialized by 1990. The naphtha (Cc-200°C boiling range) from
the H-Coal process can be severely hydrotreated to produce a feedstock
that should be capable of being steam reformed to provide synthesis gas
23 24 25
fuel for a fuel cell. ' ' For the purposes of this study, the
heavier distillate products (200°C+) of the H-Coal process are consid-
ered uneconomical to hydrocrack into naphtha and are sold as a low-
sulfur fuel oil.
IV-53
-------
COAL MINE
COAL
LIQUEFACTION
PLANT
LIQUIDS
PIPELINE
NAPHTHA
DISTRIBUTION
26-MW
FUEL CELL
POWER PLANT
ELECTRICITY
DISTRIBUTION
HEAT PUMP
FIGURE IV-15. BLOCK FLOW DIAGRAM OF SYSTEM 3
IV-54
-------
The H-Coal process has been developed by Hydrocarbon
Research Inc. (HRI) as an analog of their H-Oil hydrocracking process.
The Department of Energy, the State of Kentucky, and a consortium of the
Electric Power Research Institute, Ashland Oil, and Standard Oil of
Indiana are cooperating to build a 550 tonne per day (600 ton per day)
H-Coal pilot plant in Cattletsburg, Kentucky.
The size of the plant chosen for analysis is one that
o
produces 7,630 m (48,000 barrels) per day of mixed distillate prod-
ucts and consumes 22,300 tonnes (24,600 tons) per day or 7.3 million
tonnes (8.1 million tons) per year of subbituminous coal. The plant
3
conceptual design is a modification of one that would produce 7,950 m
(50,000 barrels) per day of distillate fuel oil. That plant is de-
scribed in Section IV-D.
b. Process Description
The H-Coal process is illustrated in Figure IV-16. Major
stream compositions and flow rates are shown in Table IV-7 for a plant
o
that produces 7,630 m (48,000 bbl) per day of naphtha and fuel oil.
Sized coal from the mine is dried, crushed, and ground to less than 60
mesh. The ground coal is mixed with an equal weight of a recycled oil
(distillation range: 345-525°C) to form a slurry which is then pumped
to 20,260 kPa (200 atm) pressure, mixed with a recycled hydrogen stream,
and then heated to 455°C (850°F). The slurry then enters the bottom
of the ebullating bed reactors, where the coal is hydrogenated and par-
tially liquefied. The reactors contain a bed of fluidized cylindrical
catalyst. Catalyst fluidization is maintained by the upward flow of
oil, coal, and hydrogen. Small fractions of catalyst may be withdrawn
for regeneration. Generally, the longer the coal-oil slurry is held in
the reactor, the lighter and more nearly hydrogen-saturated the products
become, and the greater the consumption of hydrogen. Significant frac-
tions of the sulfur and nitrogen in the coal are converted to H9S and
ammonia. Reactor products, gases, oils, ash, and unconverted coal are
cooled and separated in a series of flash drums and distillation
IV-55
-------
s
-NH3
COAL TO PLANT FUEL
H2S
NH
RECOVERY
SLURRY SOLVENT RECYCLE
CO2 VENT
STEAM
495°C+ INCL.
CHAR + ASH
H2S
+ NH-j
TO RECOVERY
PARTIAL
OXIDATION
HYDROGEN
PLANT
AIR
©
ASH
-*• AND CHAR
TO DISPOSAL
12,600 m3/day
OXYGEN
PLANT
HYDROGENATION
REACTORS
H2
H2
DISTILLATION
CO2 VENT
STEAM
REFORMER
HYDROGEN
PLANT
-C4 TO H2 PLANT
..TO PLANT FUEL
495°C+ INCL.
ASH + CHAR
TO P.O.
H2 PLANT
STEAM
C-4 FROM
DISTILLATION
(FUEL + FEED)
200-345° C
2860 m3/day
. 345-495° C
1000 m3/dav
3820
200°C
m3/day
HYDROTREATING
3820 m3/day
HYDROTREATED
> NAPHTHA
FIGURE IV-16. H-COAL PROCESS FLOW DIAGRAM
-------
Table IV-7
COMPOSITION OF MAJOR STREAMS IN H-COAL PROCESS
Stream Number
Component Quantity 123 456789 10 11
Raw cnal 10 lc»/hr Qln IfiS — — _ _ _ _
* 3
H£103nm3/hr)103kg/hr — — — (280)(195) — ~ — — 5.8 —
Cx 103 kg/hr — - - - - - - - 29
C2 103 kg/hr — — — — — — — — 35
C3 103 kg/hr — — — ~ — — — — 32
C4 103 kg/hr — — — — — — — — 31
C5-200 °C 103 kg/hr — — — — — — — — — 240
200-345 °C 10 3 kg/hr — — — — — — — — — — 220
345-495 °C 103 kg/hr — — — — — — — — —
AQS °P 4. 1O Vo/tll- -- — — _ -- _ _
Ammonia tonnes/day — — — — — -•- • •• 120
12 13 14 15 16
— 0.58 (19)
2.9
3.5
3.2
3.1
— 240
—
QC __ _ __
O J
1,300 —
MAF = Moisture and Ash Free
-------
columns. The principal liquid products (boiling between 27 and 495 C)
are produced as sidecuts in the distillation columns. A list of the
components of those liquid products is presented in Table IV-8.
The separation of solids from viscous liquid residues has
proved to be very difficult. Many separation methods are under inves-
tigation, including pyrolysis, filtration, distillation, and solvent
separation. None has yet been proven clearly superior. In the flow
scheme shown in Figure IV-16, separation of solids from liquids is ac-
complished by distillation and the entire 495°C+ (925°F+) vacuum
bottoms stream is sent to the partial oxidation hydrogen plant where
most of the carbon is gasified and the ash is melted to a slag.
Table IV-8
PROPERTIES OF H-COAL DISTILLATE FUEL OILS
Distillation Range
Elemental
analysis (wt%)
Carbon
Hydrogen
Oxygen
Sulfur
Nitrogen
C5-200°C
(C5-400
200-345 °C
(400-650 °F)
345-495 °C
(650-925 °F)
84.7
13.5
1.6
0.08
0.15
86.4
11.0
2.2
0.11
0.23
88.1
8.0
3.2
0.2
0.5
C5-200°C
After
Hydrotreating
85.4
14.6
0.04
10 ppm
30 ppm
Hydrocarbon
type analysis
Paraffins
Naphthenes
Aromatic s
20
64
16
20
47
33
10
30
60
20
77
3
Source: Reference 23 (except hydrotreating data).
IV-58
-------
o
A large amount of hydrogen (623 nm per tonne or 20,000
scf per ton of coal) must be produced to satisfy the requirements of the
H-Coal process. For the process envisioned in Figure IV-16, the product
gases lighter than pentane (C5> are desulfurized and steam-reformed to
produce hydrogen in addition to that supplied by the partial oxidation
plant. The cut point on the distillation column (495°C) is adjusted
to vary the partial oxidation plant feed and thus balance the total hy-
drogen requirement.
A large air separation plant is needed to supply oxygen
to the partial oxidation hydrogen plant. The partial oxidation plant
also provides a convenient disposal of difficult-to-separate tars and
phenolic compounds produced in the H-Coal process. Some early tests
have indicated that H-Coal process residues will provide a satisfactory
feed to the partial oxidation plant.
In this system, coal is liquefied in a mine-mouth H-Coal
plant. A portion of the coal liquid (the naphtha) is sent by pipeline
to 26-MW fuel-cell stations at dispersed locations. At the fuel-cell
stations, the naphtha must be steam-reformed to provide a synthesis gas
(CO and H.) fuel for the anode of the fuel cell. If the naphtha is to
26
be steam-reformable, it must have a low sulfur content (10 ppm maximum)
to avoid poisoning the nickel reformer catalyst. The H-Coal naphtha
product (C - 200°C) will not satisfy the maximum sulfur
specification without further processing to reduce its sulfur content.
One effective method of reducing the naphtha sulfur con-
tent is to hydrotreat catalytically only the naphtha portion of the
H-Coal product. Hydrotreating has been developed for use in the petro-
leum refining industry and is simple conceptually. Liquid feed (naph-
tha) containing sulfur compounds is combined with hydrogen and passed
over a hot (370 - 425°C), pressurized (10,300 kPa or 1500 psig) cata-
lyst. Most of the sulfur is converted to H2S, which is separated from
the clean product naphtha by distillation and acid gas absorption.
Properties of the H-Coal naphtha after hydrotreating are shown in Table
IV-8. The equipment required for naphtha desulfurization constitutes a
IV-59
-------
major processing facility. It is most reasonable to locate the naphtha
desulfurizer at the mine-mouth H-Coal plant where the required hydrogen
supply could be produced incrementally in the partial oxidation hydrogen
plant. In this case, the extra feed for the partial oxidation plant is
supplied by lowering the end point of the H-Coal distillate product from
525°C to 495°C (975°F to 925°F), which reduces the distillate
output from the H-Coal plant by 3 weight percent. The changes required
to desulfurize the naphtha will reduce the overall efficiency by 2% down
to 64%.
An alternative to hydrotreating the naphtha before
steam-reforming would be to employ a reforming process that could
tolerate a sour (sulfur-bearing) feedstock. Such a process, called
27
autothermal reforming , is under development for use in fuel-cell
power plants. The principal conceptual difference between normal
(thermal) reforming and autothermal reforming is the manner in which the
heat for the endothermic reaction is supplied. In thermal reforming,
the catalyst is packed in specially manufactured tubes built into a
furnace. The furnace supplies the heat of reaction for the steam-
reforming. As the sulfur content of the feed increases, it must be
reformed at higher temperatures. Furnace tube metallurgy limits the use
of higher temperatures and therefore limits feed sulfur content.
Autothermal reforming generates the endothermic heat of
reaction by injecting oxygen into the catalyst bed and combusting some
of the hydrocarbon feed. Because the heat is generated inside the cata-
lyst bed, the walls can be insulated. Thus, a given metallurgy can sup-
port higher reaction temperatures and more feed sulfur.
The product from the two types of reformers will differ
in one important component — sulfur. The thermal steam-reformer has no
product sulfur and may be sent to the fuel cell anode without clean-up.
The autothermal reforming product contains too much H s to be pro-
cessed directly in a molten carbonate or possibly phosphoric acid
anode. Some form of H2s removal (for example, amine scrubbing) would
IV-60
-------
be necessary. This requirement for a sulfur removal system eliminates
autothermal reforming from this study. The fuel-cell systems studied
here assume a dispersed location, and H_S scrubbing systems that
accompany autothermal reforming would not be compatible with dispersed
locations.
3. Liquids Pipeline
Unlike the case for SNG, petroleum pipelines currently connect
the Powder River Basin with the Midwest. Two crude oil pipelines
originate in southeastern Wyoming and pass near Kansas City. One of
them might be used for shipments of synthetic petroleum products at some
later time if the production of crude oil declines in eastern Wyoming.
However, such an assumption would be speculative. To place the analysis
of System 3 on an equal footing with that of System 2, we assume that a
new pipeline must be constructed to transport coal liquefaction products
from plants located in the Powder River Basin.
A pipeline 51 cm (20 in.) in diameter is sufficient to
o
transport about 32,000 m (200,000 bbl) per day of mixed petroleum
products. A length of 1,300 km (800 mi) is required to serve the
Omaha-Des Moines-Kansas City region. Ten pumping stations employing
multistage centrifugal pumps powered by diesel engines are required over
the length of the pipeline. Unlike the situation with the natural gas
pipeline, we assume here that diesel fuel would be purchased exter-
nally. The pumping stations maintain pipeline pressures at 5,600 to
7,900 kPa (800 to 1,100 psig) at their outlets, declining to 450 to 790
kPa (50 to 100 psig) at the entrance to the next pumping station.
Two separate liquids (e.g., naphtha and fuel oil) can be sent
through the same pipeline. At the flow velocities used in most liquid
pipelines, the liquids act as though they are in plug flow, that is,
with very little backmixing. The fuel oil is pumped through the
pipeline for some time, and then naphtha is pumped after the fuel oil.
IV-61
-------
The two liquids would flow as nearly separate "plugs" of unmixed
liquid. Some mixing, called the "cuff," occurs. Depending on the
properties of the cuff, it is usually blended with one product (in this
example, the cuff would probably be blended with the fuel oil). In some
cases, greater product separation can be achieved by inserting a
liquid-filled sphere (called a "pig") into the pipeline between the fuel
oil and the naphtha. The pig moves down the pipe helping to maintain
the separation between the naphtha and the fuel oil.
4. Distribution of Naphtha
Naphtha will be moved by tank truck from a bulk terminal to
local 26-MW fuel-cell power plants. The method of distribution and the
equipment used would be similar to that now used for distributing
petroleum products such as gasoline and light fuel oils.
The design of tank trucks is influenced by market demands and
government regulations covering the weight, size, and operating speed of
the vehicle, and type of delivery it will make. Large, full trailer
tank trucks with up to 34,000-liter (9,000-gallon) capacities are used
on long runs and for bulk deliveries.
Tank trucks are now being designed with either submerged or
bottom loading to permit greater safety, faster loading time, and vapor
recovery in the bulk terminal. Unloading rates of up to 32 liters/sec
(500 gal/min) are possible with gravity discharge systems, and on-board
\
pumps permit unloading rates up to 64 liters/sec (1,000 gal/min). Vapor
recovery is now required when either the storage tank or the truck tank
is being filled. Equally stringent operating procedures are expected to
be required for naphtha fuel-cell fuel.
With a load factor of 35%, a 26-MW fuel-cell power plant will
consume 49,200 liters (13,000 gallons) of naphtha per day. An on-site
storage capacity of several days' supply, or about 150,000 liters
IV-62
-------
(54,000 gallons), would probably be required. To maintain the power
plant's fuel supply, a 34,000-liter tank truck would deliver fuel about
ten times per week.
The storage of naphtha would most likely be in underground
tanks, in much the same way that gasoline is stored at service
stations. Volumes of fuel as high as 150,000 liters are routinely
stored in such facilities.
5. 26-MW Fuel-Cell Power Plant
System 3 also uses dispersed fuel-cell power plants with 26-MW
nominal output to provide electric power to residences. Here,
coal-derived naphtha is the fuel, rather than SNG as in System 2. As a
result, a modified fuel conditioning section is required to generate
H_ and CO feed for the fuel cell. Once again, advanced molten
carbonate fuel-cell technology was selected for the power plant, based
on higher projected efficiency levels.
a. System Description
A flow plan for the System 3 power plant is shown in
Figure IV-17- The plan includes provision for desulfurizing the
coal-naphtha fuel to prevent poisoning of the steam-reforming catalyst.
Target heat rate for the system was 7,910 kJ/kWh (7,500 Btu/kWh) coupled
with water-conservative operation. Again, moderately complex heat
integration is required.
Naphtha feed (Stream 1) and recycled water (Stream 5) are
fed through a series of preheaters (E-l, E-2, and E-3) so that they
enter the reformer at 815°C (1,500°F). Although these streams are
shown as segregated, in practice they would be mixed after naphtha
vaporization to suppress carbon deposition.
IV-63
-------
NAPHTHA
AIR
FIGURE IV-17. BLOCK FLOW DIAGRAM FOR 26MW FUEL CELL POWER PLANT (NAPHTHA FUEL)
-------
The reformer operates at a 3:1 ratio of steam to carbon.
At these temperatures and RJO/C conditions given above, the methane
slip equals 0.7% of the reformer carbon feed. Most of the reformer
effluent (Stream 9) is fed to the anode (Stream 10) after being cooled
in E-2, where it preheats the reformer feeds.
Naphtha must first be hydrodesulfurized to transform the
sulfur compounds to H S, which are then removed with the use of a ZnO
adsorbent. The desulfurization procedure is shown in Figure IV-17.
Stream la is withdrawn from E-l and sent to the hydrodesulfurization
(HDS) reactor. There, it reacts with hydrogen (Stream 36), producing
H S, which is removed by the ZnO. Stream Ib is then returned to E-l
with 0.2 ppm S, which is low enough to protect the reformer catalyst
from sulfur poisoning. The hydrogen stream (Stream 36) is obtained by
diverting about 1.5% of Stream 9 (Stream 34) into a series of coolers
(E-3 »~ Stream 36)and a shift converter (Sh—»-Stream 36) to shift CO
to H and CO . The final hydrogen stream (Stream 36) is mostly H
and HO. The small amounts of CH, , CO, and CO- in Stream 36 were
neglected in further analyses. In the molten carbonate fuel cell, 75%
of the combined H and CO in Stream 10 are reacted. (CH, remains
inert at this temperature.) The utilized H and CO are oxidized to
H_0 and C02> and a stoichiometric amount of CO., is transferred to
the anode via the electrolyte. Unused anode fuel, consisting of CH,,
CO and H,. (Stream 11), is preheated in E-4 and combined (Stream 11')
in the burner with 10% excess air (Stream 14), which has been preheated
in E-4 and E-5. The burner gases are catalytically ignited, producing
an adiabatic flame temperature of 1,224°C (2,236°F). The burner
gases (Stream 15) supply heat to the catalytic reformer, preheat (Stream
16) the reformer feeds in E-3, preheat (Stream 17) the burner feeds in
E-4, and again preheat (Stream 18) the cold reformer feeds in E-l.
Finally, the cooled burner gases (Stream 19) are further cooled in E-7
by air (Stream 22) until enough water is condensed to maintain a net
H»0 balance for the system. Cooler gas and condensate (Stream 20) are
separated in the knockout drum into a recycled H90 stream (Stream 5)
and a saturated gas (Stream 21).
IV-65
-------
Part of the cooling air (Stream 25) from E-7 is mixed
with the CO -rich burner exhaust (Stream 21) for use as cathode
make-up feed. That mixture (Stream 26) is preheated in E-6 and blended
with a recycled cathode exhaust stream (Stream 31) to provide cathode
feed (Stream 28). In the cathode, 1/2 0 and CO react stoichio-
metrically into the electrolyte. Half of the 0 fed to the cathode is
reacted per pass. The cathode exhaust (Stream 29) is used to preheat
the burner air (Stream 12) in E-5. Seventy percent of the cathode
exhaust is recycled (Stream 31) and mixed with Stream 27 to provide
cathode feed (Stream 28). The remaining cathode exhaust (Stream 32)
preheats the cathode make-up feed (Stream 26) and is released to the
atmosphere (Stream 33).
This design meets the goals of heat rate and HO con-
servation, but various modifications of the sytem could also meet the
same design goals. All such designs still require fairly elaborate
thermal integration.
b. System Design Basis
The general design bases for integrated molten carbonate
fuel-cell systems were discussed earlier in Section IV-B.
The properties of the hydrotreated H-Coal naphtha product
to be used in the fuel-cell power plant were presented in Table IV-8.
The thermal properties of the naphtha were estimated as:
o Heat of Combustion (HHV): 46.7 MJ/kg (20,100 Btu/lb).
o Heat of Vaporization: 325 kj/kg (140 Btu/lb).
o Vapor Heat Capacity at 16°C (60°F): 1.65 kJ/kg-°C
(0.395 Btu/lb-op).
IV-66
-------
The values reflect a weighted average of similar naphthenes and
paraffins. Those estimates were used to calculate naphtha enthalpies,
21
consistent with the Girdler data book bases.
The use of coal-derived naphtha fuel imposes an
additional system design constraint. Sulfur compounds in the naphtha
can gradually poison both the reformer catalyst and the fuel cell
electrodes. Hydrodesulfurization (HDS) and ZnO treatment of the feed
naphtha is necessary to reduce the sulfur content below 0.2 ppm. The
hydrogen recycle stream used in the HDS reactor is passed through a
small shift conversion section to lower the CO content to levels that
will not poison the HDS catalyst (Ni-Mo on alumina).
Naphtha fuel delivered to the power plant site has
already undergone extensive desulfurization processing. We assume that
the remaining refractory organic sulfur compounds can be hydrotreated
successfully on-site, but laboratory verification of this assumption is
required, using specific coal-naphtha feed stocks.
As indicated, different steam reforming conditions were
selected for the naphtha feed (815°C, H.O/C = 3), compared with the
°
SNG feed (760C, EJO/C = 4). This change does not reflect the
relative ease of steam reforming; rather, it reflects an evolution in
process design during this study. Flow plan optimization requires the
iterative assessment of varying fuel-cell design voltages, reformer
operating conditions, and system thermal integration. These analyses
must be carried through to the cost estimation stage. Such optimization
was beyond the scope of this study.
Acceptable methane slip values can be achieved by
increasing the reformer temperature and/or increasing the steam/carbon
ratio. Lower steam diluent concentration increases anode performance in
the fuel cell. However, this effect is counterbalanced by the
favorable effect of high partial pressure of the water on the
IV-67
-------
equilibrium conversion of CO to H within the anode compartment. The
size and cost of the water recovery condenser is also affected.
Further, optimal designs for SNG and naphtha fuels differ somewhat,
based on differing carbon/hydrogen ratios in the feed, and the impact of
resulting CO partial pressures on cathode performance for typical
process flow integration. Again, considerable opportunity exists for
future optimization.
Lastly, some comments are in order to justify the
selection of steam reforming as the fuel conditioning process.
Autothermal reforming was considered briefly as a possible alternative.
This process can be used with relatively heavy feedstocks, beyond the
naphtha boiling range. However, its use in the naphtha-fueled power
plant did not appear promising, based on the following considerations:
o Lower anode performance is expected, due to the presence of
air-derived nitrogen diluent in the anode feed. Also, the
fuel conversion efficiency of the autothermal process will be
lower than for steam reforming.
o The combustion value of the spent anode fuel, containing
unused H2 and CO reactants, cannot be used effectively
because the autothermal reformer has low fuel firing
requirements. Furthermore, the 26-MW power plant is probably
too small for cost-effective addition of a gas turbine or
steam bottoming cycle for generating additional electrical
energy from the spent anode fuel.
c. System Operating Characteristics
The final system material and heat balance at full load
operation is summarized in Table IV-9. Net power output is estimated to
be 25.6 MW, with a system heat rate of 7,720 kJ/kWh (7,315 Btu/kWh) (HHV
basis).
Corresponding fuel-cell performance was estimated at
0.78 V/cell at an average current density of 165 mA/cm2. Again, near
IV-68
-------
Table IV-9
PROCESS FLOW STREAMS FOR 26-MW FUEL-CELL POWER PLANT (NAPHTHA)
vO
Temperature
Stream
1
la
Ib
2
3
4
5
6
7
8
9
9'
10
11
11'
12
13
14
15
16
17
18
19
20
°C
15.6
349
341
462
689
816
66
462
689
816
816
816
593
704
760
15.6
649
760
1,224
870
816
747
234
66
(°F)
(60)
(660)
(645)
(863)
(1,273)
(1,500)
(150)
(863)
(1,273)
(1,500)
(1,500)
(1,500)
(1,100)
(1,300)
(1,400)
( 60)
(1,200)
(1,400)
(2,236)
(1,598)
(1,500)
(1,376)
(454)
(150)
Enthalpy
(GJ/hr)
-7.52
-2.70
-2.88
-1.07
2.77
5.15
-244.4
-193.4
-185.3
-189.6
-129.5
-127.6
-139.6
-534.9
-528.6
3.89
16.6
-19.0
-510.4
-556.2
-563.3
-571.3
-628.9
-683.8
Flow Rate
3
Flow Rate (10 g-moles/hr) Naphtha
H2
—
17.91
17.91
17.91
17.91
—
—
—
—
739.3
728.4
728.4
182.1
182.1
—
—
—
—
—
—
~
—
—
H20(
__
—
1
1
1
1
—
906
906
906
485
478
478
1,024
1,024
6
6
6
1,217
1,217
1,217
1,217
1,217
311
g) CH4 CO
— __
__
.36
.36
.36
.36
—
. 1
.1
. 1
.2 2.14 179.8
.5 2.11 177.2
.5 2.11 177.2
.4 2.11 44.30
.4 2.11 44.30
.65
.65
.65
.3
.3
.3
.3
.3
2 — —
co2
__
—
—
—
—
—
—
—
—
—
121.3
119.5
119.5
931.6
931.6
—
—
—
978.0
978.0
978.0
978.0
978.0
978.0
°2
__
—
—
—
—
—
—
—
—
—
~
—
--
~
—
129.2
129.2
129.2
11.74
11.74
11.74
11.74
11.74
11.74
N2 H2°(1) (kg/hr)
4,256 kg/hr
4,256 kg/hr
4,256 kg/hr
4,256 kg/hr
4,256 kg/hr
4,256 kg/hr
906.1
—
__
__
__
__
__
—
__
516.6
516.6
516.6
516.6
516.6
516.6
516.6
516.6
516.6 906.1
Total
4,256
4,256
4,256 + 14.74
4,256 + 14.74
4,256 + 14.74
4,256 + 14.74
542.9
542.9
906.1
906.1
1,527.8
1,505.3
1,505.3
2,184.5
2,184.5
652.4
652.4
652.4
2,723.7
2,723.7
2,723.7
2,723.7
2,723.7
2,723.7
(Continued)
Stream
1
la
Ib
2
3
4
5
6
7
8
9
9'
10
11
11'
12
13
14
15
16
17
18
19
20
*The unit for total flow rate for Streams 1 and la is kg/hr; the unit for Streams Ib, 2, 3, and 4 is kg/hr + 10 g-moles/hr;
the unit for Streams 5 - 36 is 10' g-moles/hr.
-------
Continued
TABLE IV-9
PROCESS FLOW STREAMS FOR 26-MW FUEL-CELL POWER PLANT (NAPHTHA)
Stream
21
22
23 "-
24
25
26
27
< 28
3 29
30
31
32
33
34
35
36
Temperature
°C (°F)
66
15.6
66
66
66
66
327
541
704
668
668
668
323
816
349
232
(150)
(60)
(150)
(150)
(150)
(150)
(620)
(1,006)
(1,300)
(1,235)
(1,235)
(1,235)
( 613)
(1,500)
(660)
(450)
Enthalpy
(GJ/hr)
439.4
199.9
254.7
238.6
16.1
-423.2
-388.3
-640.2
-347.2
-359.9
-251.9
-108.0
-142.9
-1.91
-2.28
-2.46
Flow Rate (10 g-moles/hr)
H2 H20(g) CH4
311.
382.
382.
360.
22.
— 333.
333.
1,111.
1,111.
1,111.
777.
— 333.
333.
10.89 7.
10.89 7.
13.37 4.
2
8
8
6
13
3
3
1
1
1 — •
8
3
3
15 0.03
15 0.03
67 0.03
CO CO
978.0
__
__
__
— •
978.0
978.0
1,675.2
996.0
996.0
697.2
298.8
298.8
2.65 1.79
2.65 1.79
0.17 4.27
°2
11.74
7,432.0
7,432.0
7,002.3
429.8
441.5
441.5
679.2
339.6
339.6
237.7
101.9
101.9
~
—
—
N2
516
29,728
29,728
28,008
1,719
2,235
2,235
7,452
7,452
7,452
5,216
2,235
2,235
—
—
—
H20(l)
.6
—
—
—
.0
.6
.6
1 ™
.1
.1
.4
.6
.6
—
—
—
Flow Rate
Naphtha
(kg/hr) Total*
1,817.6
37,543
37,543
35,372
2,170.9
3,988.5
3,988.5
10,918
9,898.8
9,898.8
6,929.1
2,969.6
2,969.6
22.51
22.51
22.51
Stream
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
The unit for total flow rate for Streams 1 and la ia kg/hr; the unit for Streams Ib, 2, 3, and 4 is kg/hr + 10 g-moles/hr;
the unit for Streams 5 - 36 is 10^ g-moles/hr.
-------
atmospheric pressure operation was assumed. The base case performance
(power density) of the System 3 fuel cell was projected to be higher
than that obtained for the System 2 power plant. Selection of a some-
what lower design voltage is partly responsible for this increase, as
noted in the discussion below. Another reason is the slightly more fa-
vorable fuel cell inlet reactant concentrations calculated for System 3:
Inlet Reactant Composition, mole fraction
System 2 (SNG)
Anode
(H2+CO)
0.560
:ha) 0.602
(02)
0.060
0.062
Cathode
(C02)
0.127
0.153
These differences reflect the impact of the higher carbon/hydrogen ratio
of the primary fuels (naphtha versus SNG) as well as evolving selection
of optimal reformer operating conditions and flow stream integration.
As discussed earlier, alternative combinations of process
variables would also achieve the efficiency and water-balance goals. In
this case, even lower design voltages could probably have been assumed.
The impact of lower voltage operation on cell current density is large,
resulting in substantially increased power density and lower fuel cell
costs. A major reason for this sensitive current density response is
the optimistic value used for cell resistance. Thin electrolyte tiles
with low ionic resistance were assumed, based on expected improvements
in cell technology. The impact of these factors on projected system
cost is discussed in Section VII-K.
d. Conceptual System Design
The conceptual design of the System 3 power plant is
similar to that developed for System 2. Power plant equipment, listed
IV-71
-------
in Table IV-10, would be assembled into modular, shippable units. Major
differences between the two conceptual designs are as follows:
o The steam reforming reactor contains a different catalyst,
Girdler G-91C. This potassium-activated nickel on alumina
catalyst was developed by Girdler for naphtha reforming. The
vendor claims improved activity and lower cost compared with
previous available catalysts.
o The reformer package again contains the burner, reformer
section, and heat exchangers E-l through E-4. The surface
areas of the heat exchangers, however, are larger than those
in the System 2 module. Thus, the overall size, weight, and
cost of the package are greater. It may be necessary to use a
larger number of reformer units (perhaps one for each
fuel-cell trailer or a total of eight) to avoid exceeding the
size and weight limit for a shippable unit. For this study,
however, we assume that four reformer units are used.
o The fuel cell operates at a higher power density, resulting in
a smaller active area requirement per cell and lower fuel cell
cost.
o The components of the HDS loop can be fitted into the
equipment module envelope. Additional space was available,
based on a reduction in the length of exchanger E-6. The HDS
reactor bed was sized assuming a design space velocity of
1,000 m-* output/hr-m^ catalyst.
o The arrangement of modules is nearly identical to that shown
in Figure IV-12 for the SNG-fueled power plant. The overall
plant size is 58.5 x 29 m (192 x 128 ft), somewhat larger than
the SNG-fueled plant, because of the larger air fin condenser
E-7.
6. Distribution of Electricity
See Section IV-B for the complete description of the
distribution of electricity from a 26-MW fuel-cell power plant.
7. Heat Pump
See Section IV-B for the complete description of the use of
heat pumps for residential heating and cooling.
IV-72
-------
Table IV-10
EQUIPMENT LIST FOR 26-MW FUEL-CELL POWER PLANT (NAPHTHA)
-J
U>
Unit
Molten Carbonate Fuel Cell
Heat Exchangers
E-l
E-2
E-3
E-4
E-5
E-6
E-7
E-8
Reformer Section
Reformer Catalyst
Burner
Shift High-Temp. Catalyst
Shift Low-Temp. Catalyst
Shift Intercooler
Hydrodesulfurization Catalyst
ZnO Bed
Knockout Drum
Condensate Pump
Blowers
B-l
B-2
B-3
B-4
B-5
B-6
Capacity
28.4 MW D C
57.6 GJ/hr
12.0 GJ/hr
7.1 GJ/hr
8.0 GJ/hr
12.7 GJ/hr
34.9 GJ/hr
54.9 GJ/hr
0.37 GJ/hr
45.9 GJ/hr
740,020 g-mole Hj/hr
2,183,700 g-mole/hr
22,510 g-mole H2/hr
0.19 GJ/hr
4,256 kg naphtha/hr
1.02 kg S/day
906,200 g-mole H2o/hr
4.5 liter/sec.
249 m3/min
14,000 m3/min
2,580 m3/min
679 m3/min
821 m3/min
5.5 m3/min
Size
22.1 x 103 m2
1.40 (x 103 m2)
0.35 (x 103 m2)
0.45 (x 103 m2)
1.01 (x 103 m2)
0.26 (x 103 m2)
1.15 (x 103 m2)
2.40 (x 103 m2)
0.009 (x 103 m2)
0.91 (x 103 m2)
21.3 m3
179 GJ/hr
0.77 m3
0.35 m3
0.68 m3
1.20 m3
1.567 m3
8.05 m3
—
—
—
—
—
—
—
Modular Size
2.77 x 103 m2
0.35 (x 103 m2)
0.088 (x 103 m2)
0.11 (x 103 m2)
0.25 (x 103 m2)
0.065 (x 103 m2)
0.29 (x 103 m2)
0.60 (x 103 m2)
0.003 (x 103 m2)
0.23 (x 103 m2)
5.33 m3
44.6 GJ/hr
0.19 m3
0.087 m3
0.17 m3
0.30 m3
0.40 m3
2.01 m3
1.1 liter/sec
62.3 m /min
3,510 m3/min
651 m /min
170 m /min
205 m3/min
1.4 m /min
Dimensions
Fuel-Cell Trailer
3.1 :t 7.3 x 3.7m
Included in
Reformer Package
3.8 m dia x 7.3 m high
122 cm dia x 3.7 m long
122 cm dia x 7.0 m long
8.5 x 9.1 x 4.6 m high
0.30 x 0.30 x 0.61 m
Included in Reformer Package
-
Included in Reformer Package
0.45 m dia x 1.7 m high
0.30 m dia x 1.1 m high
0.15 m dia x 0.30 m long
-
0.61 m dia x 2.1 m long
1.3 m dia x 2.1 m long
0.61 x 0.61 x 1.2 m long
0.91 x 0.91 x 0.91 m
Part of E-7
1.5 x 1.5 x 1.8 m
1.2 x 1.2 x 1.2 m
1.2 x 1.2 x 1.2 m
0.30 x 0.30 x 0.30 m
Materials
UTC Advanced Design
304 SS
309 SS
304 SS
304 SS
304 SS
304 SS
C.S. Finned
309 SS Finned
HK40 Cr-Ni Tubes
Girdler G-91C
C.S. and Refractory
Girdler G-3A
Girdler G-66
C S
Girdler G-51C
Girdler G-72
Galvanized
C.S.
C.S.
C.S.
304 SS
C.S.
C.S.
C.S.
-------
D. System 4
A block flow diagram of System 4 is shown in Figure IV-18. The
components of the system are described below.
1. Coal Mine
See Section IV-A for the complete description of a surface
coal mine located in the Powder River Basin.
2. Coal Liquefaction Plant
The H-Coal liquefaction process was described in Section
IV-C. The product of interest in that discussion was hydrotreated
naphtha, which was to be used as fuel for the 26-MW fuel cell. In
System 4, the required product is a low-sulfur distillate fuel for a
combined-cycle power plant. The flow diagram for a facility that
3
produces 7,950 m (50,000 bbl) per day of such a product is shown in
Figure IV-19; the stream compositions are shown in Table IV-11. The
principal difference between the facility shown in Figure IV-19 and that
shown in Figure IV-16 is that the C -200°C naphtha stream is not
further hydrotreated. Rather, the entire H-Coal distillate product is
sold as-is, for use as turbine fuel in a combined-cycle power plant.
The quality inspection for the fuel was shown in Table IV-8.
The major process difference between the two facilities is
that when fuel oil is the required product, the end point of the
distillate product is raised from 495°C (925°F) to 525°C
(975 F), raising the distillate product output by 3%. The increase in
the end point reduces the feed to the partial oxidation plant, in
conformance with the reduced hydrogen requirement of the process.
IV-74
-------
COAL MINE
COAL
LIQUEFACTION
PLANT
PETROLEUM
PRODUCTS
PIPELINE
FUEL OIL
DISTRIBUTION
COMBINED CYCLE
POWER PLANT
ELECTRICITY
DISTRIBUTION
HEAT PUMP
FIGURE IV-18. BLOCK FLOW DIAGRAM OF SYSTEM 4
IV-75
-------
©
RAW COAL
COAL TO PLANT FUEL
SLURRY SOLVENT RECYCLE
FEED
CO0 VENT
f
STEAM
525-C+ INCL.
CHAR + ASH
H2S
NH-
TO RECOVERY
PARTIAL
OXIDATION
HYDROGEN
PLANT
AIR
©
ASH AND
CHAR TO
DISPOSAL
OXYGEN
PLANT
12,600 rtvVday
HYDROGENATION
REACTORS
H2
C4 TO H2 PLANT
PLANT FUEL
CO2 VENT
STEAM
REFORMER
HYDROGEN
PLANT
525eC+ INCL.
ASH + CHAR
TO P.O.
H2 PLANT
• STEAM
C-4 FROM
' DISTILLATION
(FUEL + FEED)
C5- 200° C
3820 m3/dav
200-345° C
2860 m3/day
345-525° C
1210 m3/day
FIGURE IV-19. H-COAL PROCESS FLOW DIAGRAM
-------
Table IV-ll
COMPOSITION OF MAJOR STREAMS IN H-COAL PROCESS
Stream Number
Component
Raw Coal
MAF* Coal
H£l03nm3/hi
Cl
C2
C3
C4
C5-200 °C
200-345 °C
345-525 °C
525 °C +
Char
Ash
Oxygen
Sulfur
Ammonia
Quantity 123 456789 10 11 12
3
103 kg/hr — — — — — — — — 35
103 kg/hr — — — ~ — — — — 32
103 kg/hr — — — — — — — — — 240
3
1 n lr» />»«• — — — — — — — — — i nn
3
i n v» /tiw — -— — — _ — _ — —
tonnes/day — — — — — — — 120
13 14
2.9
3.5
3.2
3.1
120 —
1,300 —
1,100 —
—
MAP " Moisture and Ash Free
-------
Although the suitability of coal-derived fuels for use in gas
turbines has not yet been demonstrated, an H-Coal distillate fuel with
the properties shown in Table IV-8 is expected to meet most
28
specifications for turbine fuels, partly because the fuel contains
no 525°C+ (975°F+) fraction, and thus has a relatively high H/C
ratio.
3. Liquids Pipeline
See Section IV-C for a complete description of an interstate
pipeline suitable for transporting a combination of liquid petroleum
products.
4. Fuel Distribution
Because of the large volume of fuel required for the
combined-cycle power plant, the most effective method of supply would
probably be railroad tank cars. For the 270-MW power plant described in
o
the following section, 430 m (2,700 bbl) of distillate fuel oil would
be consumed daily, assuming a load factor of 35%. If the power plant is
located a significant distance from the pipeline terminus, shipments of
fuel oil by unit train from the bulk terminal to the power plant are an
effective means of supply.
A unit train consisting of 80 tank cars, each with 37,850
liter (10,000 gallon) capacity, could supply the power plant
requirements with one weekly delivery. To ensure against disruptions in
the supply, a significant on-site storage capacity is required. For
large, centralized power plants, that requirement is typically met by
above-ground steel storage tanks. To maintain a 30-day fuel supply
would require a storage capacity of 13,000 m3 (3.5 million gal).
IV-78
-------
5. Combined-Cycle Power Plant
a. Background
Conceptually, a combined-cycle power plant is similar to
a fossil-fired steam electric plant, with the addition of a device for
improving the process efficiency. The device is a gas turbine-generator
set that extracts work (electricity) from the combustion gases before
they go to the steam boiler-steam turbine-generator system. The gas
turbine enables the system of combined cycles to take advantage of
higher combustion temperatures than would be practical with a steam
turbine alone. That advantage results in a higher system efficiency.
Even higher efficiencies could be obtained if gas turbines can be
developed to withstand still higher inlet temperatures. The system used
as an example for this study was modeled on a system designed by
29
Westinghouse for the Energy Conversion Alternatives Study (EGAS).
The gas turbine analyzed in that study can tolerate an inlet temperature
1,380°C (2,500°F). The high inlet temperature would be made
possible by using air-cooled ceramic gas turbine blades and nozzles.
Currently, typical operating conditions for utility turbines are:
o Turbine inlet temperature between 980 and 1,090°C
(1,800 and 2,000°F).
o Compression discharge pressure ratio (compressor
discharge/ambient pressure) 8:1 to 15:1.
Because there is such a wide gap between the current clean-oil-fed
turbine technology and that advocated in the Westinghouse study, some
people have recommended developing an intermediate technology that would
use residual petroleum feeds for turbines.
IV-79
-------
High combustion temperatures are not always beneficial.
More nitrogen oxides (NO ) are formed with high-temperature
X
combustion. Additional water can be injected to reduce the NO
effluents, but that also reduces the turbine inlet temperature and,
31
therefore, also reduces the process efficiency. Various NO
X
reduction methods for gas turbines are being studied, but none are
currently able to provide passable performance with coal-derived liquids
for an NO limit of 0.14 kg per GJ (0.3 Ib per million Btu) (HHV).
X
Two sources of nitrogen form NO :
o Chemically bound nitrogen in the gas turbine fuel
o Molecular nitrogen from combustion air.
Natural-gas-fueled combined-cycle plants have no
chemically bound nitrogen in the fuel and thus must cope only with
"thermally fixed" nitrogen from combustion air. However, just as the
NO emission regulations are being tightened, changes in turbine fuels
X
toward those with considerable chemically bound nitrogen content are
being mandated . H-Coal distillate is an example of such a high
nitrogen feed. Nitrogen can be removed from the H-Coal distillate by
hydrotreating, but it is not economical. Just developing the hardware
o o
that can burn the hydrogen-deficient fuel is a problem. Lowering
o o
the NO^ emission is a second-order objective, but one that must be
met if combined-cycle systems are to achieve widespread application.
b. Process Description
Figure IV-20 shows a block flow diagram of a
coal-liquid-fired combined-cycle plant. The fuel (H-Coal distillate) is
first prepared by filtering, washing with water, and centrifuging to
remove soluble salts of sodium and potassium. Magnesium fuel additives
(e.g., MgSO^-7H20 or MgO emulsion) can also be used to inhibit
vanadium attack in the turbine.34,35
IV-80
-------
EVAPORATED WATER
COMBUSTION
GASES TO STACK
HEAT
RECOVERY
STEAM
GENERATOR
®
STEAM
STEAM
TURBINE
DISTILLATE
FUEL
FUEL
PREPARATION
COMPRESSOR
AIR
WASTE
WATER
WATER FOR
NOXSUPRESSION
COMBUSTOR
COOLING AIR
SLOWDOWN
WATER
GAS
TURBINE
94 MW
GENERATOR
176MW
FIGURE IV-20. COMBINED CYCLE POWER PLANT BLOCK FLOW DIAGRAM
IV-81
-------
Air for combustion is compressed to a pressure ratio of
16:1 in the gas turbine driven compressor. Fuel and compressed air are
burned in the combustor. The 1,370 C (2,500 F) combustion gases
drive the gas turbine-generator set.
Exhaust gases from the gas turbine then enter the
non-fired heat recovery steam generators (HRSG) at 670°C (1,240°F).
High-pressure steam (16,500 kPa or 2,400 psia) is raised in the HRSGs to
drive a conventional steam turbine-generator set. Condensed exhaust
from the steam turbine is returned to the HRSGs.
Table IV-12 shows the mass flow rates for the 270-MW
combined-cycle power generating plant. Several streams are combined in
Table IV-12 to show overall flow rates. Two gas turbine-generator sets
have 91-MW capacity each. There is one HRSG for each gas turbine, and
these send combined steam flows to the single 97-MW steam
turbine-generator set.
Not shown in Table IV-12 or Figure IV-18 is a single
reheat cycle in which 350°C (660°F) steam exhausted from the first
stage of the steam turbine is routed to the HRSG for reheating to
540°C (1,000°F). The reheated steam powers the second and third
(intermediate and low pressure) stages of the steam turbine.
6. Transmission and Distribution of Electricity
See Section IV-A for the complete description of the
electricity transmission and distribution system.
7. Heat Pump
See Section IV-B for the complete description of the use of
heat pumps for residential heating and cooling.
IV-82
-------
Table IV-12
MASS FLOW RATES FOR 270-MW COMBINED-CYCLE POWER PLANT
Stream
Name and Number
1 Distillate Fuel
2 Air
3 Waste-water
4 Water for NO
x
Suppression
5 Cooling Air
6 Steam
7 Combustion Gases
8 Evaporated Water
9 Slowdown Water
Mass Flow Rate Temperature
(103 kg/hr) °C (°F)
1
42
,500
6
150 (300)
60 (16)
— —
Pressure
kPa (psia)
101 ( 14.7)
101 (14.7)
— —
22
58
250
1,550
190
180
540 (1,000) 16,500 (2,400)
670 (1,240) 107 (15.6)
100 (212) 101 (14.7)
TV-83
-------
E. System 5
A block flow diagram of System 5 is shown in Figure IV-20. The
components of the system are described below.
1. Coal Mine
See Section IV-A for the complete description of a surface
coal mine located in the Powder River Basin.
2. Coal Gasification Facility
See Section IV-A for the complete description of the Hygas
coal gasification process.
3. Gas Pipeline
See Section IV-A for the complete description of an interstate
natural gas pipeline.
4. Gas Distribution
See Section IV-A for the complete description of the natural
gas distribution system.
5. 100-kW Fuel-Cell Power Plant
This power plant uses a first-generation phosphoric acid fuel
cell as an energy converter in a total energy package. Heat recovered
IV-84
-------
COAL MINE
COAL
GASIFICATION
PLANT
GAS PIPELINE
GAS
DISTRIBUTION
100-kW
FUEL CELL
POWER PLANT
HEAT PUMP
HEAT
RECOVERY
FIGURE IV-21. BLOCK FLOW DIAGRAM OF SYSTEM 5
IV-85
-------
from the fuel cell is used to provide hot water to satisfy part of the
thermal load demand in a housing complex. The system concept is based
on published United Technologies Corporation reports, supplemented by
cell design and performance data from current phosphoric-acid fuel-cell
technology programs at Energy Research Corporation and Universal Oil
Products. Target electrical generation efficiency is in the 30 to 35%
range for expected operation at partially rated to full-rated electrical
load (100 kW).
a. System Description
The 100-kW power plant contains fuel-cell power stacks
and associated auxiliary process equipment, including a fuel conditioner
that converts SNG feed into hydrogen fuel, air supply system, waste heat
recovery system, and an inverter to convert power plant DC output to
AC. Figure IV-22 is a block diagram showing the functional arrangement
of these components.
Details of the process flow stream integration are shown
in Figure IV-23. Odorant sulfur compounds are removed from the incoming
SNG feed in a small zinc oxide guard bed. Steam is added to the SNG
feed prior to entering the reformer section, where the methane-steam
reforming reaction is carried out to produce hydrogen for the fuel
cell. The hot reformed gases are then cooled and passed through a
series of shift conversion reactors to reduce carbon monoxide content
prior to entering the fuel cell. Spent fuel from the fuel cell is used
to fire the reformer furnace. A heat exchanger network is provided to
accomplish the required reactant preheat and product cool-down.
The fuel-cell stacks are cooled with a circulating
silicone oil coolant. The hot oil is first used to generate steam for
the reformer. Major waste heat recovery is then accomplished by cooling
the oil exchanger E-7, using a hot water stream. This stream is heated
to 82 C (180 F), the design return temperature for the housing
IV-86
-------
AC POWER
INVERTER
FUEL
CONDITIONER
HOT WATER
PRODUCT
DC POWER
FUEL CELL
STACKS
FIGURE IV-22. 100-kW POWER PLANT COMPONENTS
IV-87
-------
FIGURE IV-23. HEAT INTEGRATION FOR 100-kW FUEL-CELL POWER PLANT
IV-88
-------
complex hot water system. Additional quantities of 82°C water are
produced by cooling the combined exhaust air and reformer furnace flue
gas streams in exchanger E-6. Final combined gas cooling is carried out
in air-cooled exchanger E-5. Here, condensed water is recovered and
used to generate process steam for the reformer.
Depending on the demand, situations may arise where
electrical power is required, but thermal energy (hot water) is not.
Steady-state operation of the fuel-cell system will still require re-
moval of waste heat from the circulating oil coolant and water recovery
from the exhaust gas streams. To accomplish this, an air-cooled bypass
exchanger E-8 is provided in the oil coolant loop. In addition, trim
cooler E-5 in the air loop is oversized to permit total cool-down of the
exhaust gas streams to the design dew point.
b. System Design Basis
This section discusses the rationale for the specific
design parameters selected for the 100-kW power plant study. Phosphoric
acid fuel-cell technology was used because extensive programs are
currently underway to develop this technology for total energy power
plants. Participants in these programs include United Technology
Corporation (UTC), Energy Research Corporation (ERG), Westinghouse, and
Engelhard Industries. The designs developed in this study represent a
synthesis of various features of the UTC and ERG systems.
Fuel-Cell Design — The static electrolyte fuel cell contains
multiple assemblies of cells formed by stacking electrodes sandwiched
around a silicon carbide matrix and formed bipolar plates containing
grooves for reactant distribution. Cost studies were based on cells
designed by ERC for the U.S. Army, which have been described in
considerable detail .
Platinum electrocatalysts were assumed, with a total (anode
n
plus cathode) loading of 1 mg/cm . This is somewhat higher than the
IV-89
-------
n
0.75 mg/cm loading in UTC cells, but appears consistent with current
ERG values. Data for Engelhard cells are not available.
Cell cooling is accomplished using cooling plates within the
stack. These plates contain embedded cooling tubes for circulating an
inert silicone oil coolant, based on an early UTC design concept.
Further study of this configuration revealed that the oil was not truly
inert, and if leakage occurred, subsequent corrosion products had an
adverse effect on fuel-cell performance. Current UTC designs employ a
two phase pressurized boiling water coolant, used in cells operating at
a nominal temperature of 190°C (395°F). ERG and probably Engelhard
designs, on the other hand, use circulating air as coolant. These
differences may have an impact on the heat transfer surface area
required to recover waste heat from the fuel cell. The silicone oil
design is assumed to be adequate for purposes of this study.
Fuel-Cell Operating Conditions and Performance — Fuel-cell
operating temperature and pressure were set at 170°C (338°F) and
1 atm, respectively. These conditions appear to offer a workable
compromise among several competing factors including stack performance,
life, and cost.
Fuel-cell performance improves as temperature and pressure are
raised. Current UTC designs call for large, multi-megawatt power plants
that operate at 190°C (375°F) and 640 kPa (93 psia). Improved per-
formance can result in improved power plant efficiency (if operating
voltage is increased) or in reduced power plant cost (if operating cur-
rent density, hence power density, is increased). On the other hand,
prospects for achieving target stack life, assumed to be 40,000 hours,
are reduced at these more severe conditons. Life-limiting factors in-
clude gradual loss of cathode catalyst surface area via sintering and
dissolution, and degradation of electrode structure caused by corrosion
of the carbon supports on which the platinum catalyst is dispersed. The
latter adversely affects the reactant gas-electrolyte-catalyst interface
within the electrode structure.
IV-90
-------
Previous ERG stacks, developed for the U.S Army, operate at
176 C (350 F) and recent cell evaluations using catalysts supplied
by Universal Oil Products Corporation have been conducted at 180°C.
To date, there have been no published reports of phosphoric
acid fuel-cell stack operation extending to 40,000 hours. For this
study, we selected a low operating temperature of 170°C (338°F) as
most likely to achieve that target Lifetime.
Waste heat utilization applications can also influence the
choice of operating temperature. Cogeneration schemes requiring sub-
stantial amounts of heat as low-pressure steam appear more attractive if
higher fuel-cell operating temperatures are selected. In System 5, we
assume that all waste heat is used to produce relatively low grade hot
water at 82°C. Fi
this application.
water at 82 C. Fuel-cell operation at 170 C appears adequate for
Loss of phosphoric acid electrolyte via vaporization from the
cell matrix and electrolyte structures is another life-limiting factor
controlling the choice of operating temperature. Those vaporization
losses (into the flowing air stream) increase as temperature increases.
Operation at high pressure can reduce electrolyte loss, but cost-
effective means to do this are not available (see below). Again,
operation at 170°C appears to offer a suitable compromise.
High-pressure operation improves performance. However, to
operate efficiently at elevated pressure, a coupled turbocompressor/
expander set must be used to compress ambient air and to recover energy
from the hot compressed fuel-cell exhaust air stream. Efficient devices
that accomplish those tasks at the flow rates used in the 100-kW power
plant are too costly.
Increased operating temperature reduces the extent of
reversible adsorption or "poisoning" of the platinum anode catalyst by
carbon monoxide. The latter is present in the reformed gas feed.
IV-91
-------
Operation at 170°C would tend to aggravate this poisoning effect.
Published phosphoric acid fuel-cell performance
characteristics were reviewed. Those data are shown in Figure IV-24 and
Table IV-13. As indicated, the data base is sparse. Few total cell
performance data have been published, particularly for advanced cell
designs such as those of UTC, which are usually considered proprietary.
Figure IV-24 shows initial performance data for relatively
small single cells at a variety of temperatures. Life-limiting
phenomena will cause performance degradation with time. In addition,
performance losses can be expected as cells are scaled-up to commercial
size, stacked into multicell assemblies, and used in the field. It is
not possible to predict such losses at this time.
Figure IV-24 also shows the base-case performance curve and
design point used in this study. This estimate of average cell
performance in the 100-kW power plant at end-of-life conditions (40,000
hr) is assumed to be conservative. Clearly, performance will improve as
cell designs evolve. Figure IV-24 also includes a projected design
point, used to assess the impact of advanced phosphoric fuel cell
performance.
The base case design curve was established in mid-1977.
Subsequently, published cell performance data show that improvements
have indeed occurred, particularly for cells operating at elevated
temperature (190°C) and pressure (345 kPa or 50 psia). Additional
advances are required to achieve target lifetimes.
Fuel Conditioning System — The fuel conditioning system is
used to convert the SNG feed to hydrogen fuel via the endothermic steam
reforming reaction:
CH4 + 2H20 — 4H2 + C02
IV-92
-------
0.75
<3
0.70
S
1
I
z
uj
UJ
u
0.65
0.60
0.55
BASE-CASE DESIGN POINT
(40,000 hrs)
INITIAL PERFORMANCE DATA
• UTC
• ERC
A UOP
T PROTOTECH
190(c)
•-PROJECTED
IMPROVED
PERFORMANCE
•190(1)
0.50
157(d)
NOTE: Numbers are operating temperatures in °C.
See accompanying Table for details, keyed
by letters in parentheses. I
100
200
CURRENT DENSITY - ma/cm2
300
400
FIGURE IV-24. PHOSPHORIC ACID FUEL CELL PERFORMANCE DATA
-------
Table IV - 13
ADDITIONAL DETAILS FOR FIGURE IV - 24
Data
Source
UOP
UOP
Prototech
ERC
ERC
ERC
ERC
ERC
ERC
ERC
ERC
ERC
Data Set
Indicated
in Fig. IV-23
a
b
c
d
e
f
g
h
i
j
k
1
Operating
Conditions
Temp.
180
180
190
157
150
149
177
177
190d
163
190d
190
Pressure
(kPa)a
101
101
101
101
101
101
101
101
n. a.
101d
101d
345
Total
Catalyst
Loading ,
(mgPt/cmV
2.5
1.2
0.5
4.0
4.0
4.0
0.5d
n. a.
1.0
0.5
1.0
0.75
Remarks Reference
Pure H2
Pure H2
Pure H2
H + 2% CO
Pure H2
Pure H2
n. a.
n. a.
Simulated
reformate
Reformed
natural
gas
Simulated
reformate
Simulated
reformate0
37
38
39
40
41
42
43
44
45
46
47
48
101 kPa = 14.7 psia = 1 atmosphere.
Anode + cathode.
'Realistic reactant utilizations.
Assumed, not reported.
IV-94
-------
At steam reforming conditions, the water gas shift equilibrium affects
the product gas composition:
CO + H20 — H2 + C02
Reformed gas composition was estimated using a thermodynamic calculation
procedure developed by Imperial Chemicals Industries (ICI). The
computation was modified to reflect the actual Hygas SNG composition
assumed for this study:
Component Moles/Mole CH^
CH4 1.0000
CO 0.00115
C02 0.01830
H2 0.13580
H20 0.00015
Preliminary studies were carried out to establish reasonable
reforming conditions, temperature, and steam/carbon ratio, subject to
the following constraints:
o Minimum methane slip (unconverted CH^ in product gas).
Methane is an inert gas that passes intact through the fuel
cell. The anode exhaust, containing CH^ and unreacted H2
is used as burner feed for the reformer furnace. On the other
hand, excessive methane slip reduces fuel conversion
efficiency.
o Minimum CO in the product gas. Carbon monoxide is a fuel-cell
poison, as noted earlier. Iterative procedures are required
to establish CO content of the reformer effluent, because
subsequent downstream processing in the shift reactors will
reduce CO content down to target levels. A target level of
about 1 volume percent CO in the fuel cell feed (dry gas
basis) was assumed, based on cell operation at 170°C and
discussions at the 1977 National Fuel Cell Seminar held in
Boston. Recent UTC reports indicate that higher temperature
cells (190^C) are designed to operate with about 1.7 volume
percent C0^8.
Trade-off studies were carried out to develop an internally
consistent, but not necessarily optimal, balance between those
IV-95
-------
constraints. Methane slip decreases rapidly as the reforming
temperature is increased, but CO content also increases. Methane slip
and CO content both decrease as the steam/carbon ratio increases, but
excessive H 0/C ratios involve costly water recovery heat exchangers
and increased energy consumption for steam generation.
The final integrated system heat balance resulted in a
somewhat excessive temperature for the reformer furnace flue gas
effluent. Additional optimization studies are required to lower this
temperature. Nevertheless, the system as designed provides adequate
flame temperature and temperature differences within the furnace to
carry out the required heat transfer.
Heat Recovery System — The heat recovery system was designed
to provide maximum hot water product at 82°C (180°F). We assumed
design hot water return temperature of 60°C (140°F). Design
revisions are required for alternative applications involving simul-
taneous heat recovery at several different temperatures.
c. System Operating Characteristics
The electrical and thermal operating characteristics of
the 100-kW fuel-cell power plant were evaluated at rated load and at
part load. A complete material and energy balance was carried out for
operation of the power plant at full rated electrical load. A fuel-cell
design point of 0.65 V was selected as an effective compromise between
high efficiency for electrical generation and production of waste heat.
The voltage-current performance characteristics of the fuel cell are
shown in Figure IV-25. As indicated, the operating current density is
9 9
112 raA/cm (104 A/ft ). The system material balance is summarized
in Table IV-14.
Potential problems with operational stability and system
control strategy were uncovered during the analysis. The thermal
IV-96
-------
0.75
0.70
§ 0.65
DESIGN POINT
I
_j
P
z
£
Q.
0.60
0.55
H3PO4 FUEL CELL
170°C, 1 atm
1 mg Pt/cm2 TOTAL
70% H2 UTILIZATION
50% O2 UTILIZATION
0.50
I
I
50
100 150
CURRENT DENSITY - ma/cm2
200
250
FIGURE IV-25. PHOSPHORIC ACID FUEL CELL PERFORMANCE CURVE
-------
TABLE IV-14
PROCESS CONDITIONS AND FLOW RATES FOR 100-kW FUEL-CELL POWER PLANT
(Design Basis: 100 kW-Net AC Power Output)
Temperature
°C
op
Pressure6
kPa
psia
Flow Rate
(g-moles/hr)
CH4
CO
co2
H2
H 0
SNG
Feed
24
75
152
22
1,216
1.36
21.8
166
Reformer
Steam Feed
112 700
233 1,292
152 152
22 22
— 1,216
1.36
21.8
166
3,648 3,648
High
Reformerfl T Shiftb
Product Product
700
1,292
145
21
106
634
500
3,972
2,061
375
707
138
20
106
275
857
4,331
1,702
Low
T Shift
Product >C
235
455
131
19
106
60. 8f
1,073
4,546
1,487
Fuel-Cell
Anode
Exhaust
170
338
103
15
106
60.8
1,073
1,363
810
Total
1,405 3,648 5,053 7,273 7,271 7,273 3,413
aRefonner outlet temperature = 700 °C (1,292 °F) @ 1.5 atm. Methane
approach to equilibrium = 15 °c (27 °F)j Teq = 685 °C (1,265 °F).
Girdler
Temperature, C ( F)
Space
Velocity
Shift Reactor Catalyst Inlet Outlet Average Equilibration (vol/vol/hr)
High-Temperature G-3A 316(600) 375(707) 346(654) 400(752) 562
Low-Temperature G-66B 204(400) 235(455) 220(428) 260(500) 761
Composition equivalent to fuel-cell anode feed.
70% H2 conversion.
^Pressures approximate only.
Equivalent to 0.83 vol. percent (wet basis), 1.05 vol. percent (dry basis).
IV-98
-------
Table IV-14 (Concluded)
Temperature
°C
oF
Pressure6
kPa
psia
Flow Rate
(g-moles/hr)
CH,
CO
co2
H2
H20
°2
N2
TOTAL '
Fuel Cell
Cathode
Feed8
69
157
103
15
—
—
—
364
3,182
11,972
15,518
Fuel Cell
Cathode
Exhaust
170
338
103
15
—
—
—
4,223
1,592
11,972
17,787
Furnace
Feed
249
480
103
15
106
60.8
1,073
1,363
926
1,016
3,820
8,365
Hot
Furnace Water
Exhaust Coolant Feed
833 170 60
1,532 338 140
103
15
—
__
1,239
2,500 — 3,122h
91.1
3,820
7,650 17,668L 3,122h
Hot
Water
Feed
60
140
—
~
—
—
l,788h
—
l,788h
ePressures approximate only.
850% 02 conversion, ambient air @ 32 °C (90 °F), 50% relative humidity.
hOil Coolant flow rate • 85,902 kg/hr (38,958 Ib/hr).
1Hot water feed flow rate in kg/hr.
IV-9 9
-------
recovery system requires bypass loops for effective operation and
maintenance of desired stream operating temperatures. The impact of
variable load operation on the highly integrated fuel processing section
was not ana- lyzed. However, considerable changes in steady state
stream tempera- tures could be encountered by operating a system
designed for 100-kW output at reduced load. Additional studies are
required to ensure that the fuel-cell operating temperature remains
close to the 170°C design value and that sufficient water is recovered
to satisfy steam reformer requirements.
d. Conceptual System Design
A conceptual design was prepared for the 100-kW power
plant components and assembled layout. The equipment list, showing ca-
pacity, size, and materials of each component, is given in Table IV-15.
Fuel-Cell Stack — The design of the phosphoric and fuel-cell
stacks is based on published ERG reports. ' The following design
parameters were used:
o Fuel-cell design performance - 112mA/cm2(104 A/ft^)
@ 0.65 V/cell
o Total DC power - 111 kW (gross)
o Number of stacks - 4
o Active area per cell - 0.19 m2 (2 ft^)
o Stack voltage - 133 V (DC)
The size of each fuel-cell stack, including fuel and air manifolding, is
approximately 94 x 43 x 107 cm (37 x 17 x 42 in.).
Reformer Package Design — The conceptual design of the re-
former is shown in Figure IV-26. It includes a burner at the base,
which mixes and feeds the anode exhaust gas and preheated air into a
IV-100
-------
Unit
Type
Capacity
Table IV-15
100 kW FUEL-CELL POWER PLANT
Size Dimensions
Weight (kg)
Phosphoric Acid Fuel Cell E.R.C.
Reformer Package
Reformer Catalyst
Low Temperature
Instrumentation
Base and Enclosure
Piping, Wiring, Misc.
Ill kW DC
Vertical Tube 233 MJ/hr
Total Cell Area 153 m2 4 stacks each 94x43x119 cm 1,590
Reformer Section 2.85 m
Convection Section 0.73m2 84x84x173 cm
Pellets 3980 g-mole E2 0.036 m3
Low Pressure 6130 g-moles 428 MJ/hr Included in reformer package
38 x 38 x 15 cm
10 x 15 x 28 cm
14 x 15 x 36 cm
13 x 23 x 28 cm
14 x 20 x 36 cm
25 x 38 x 56 cm
1.2 x 1.5 x 1.2 m
25cm dia x 194 cm long
16.5cm dia x 104 cm long
53 x 53 x 50 cm
81 x 76 x 86 cm
M
<
1
I-1
O
t— '
Heat Exchangers
E-1A
E-1B
E-2A
E-2B
E-3
E-4
E-5
E-6
E-7
E-8
Blower B-l
Pumps P-l
P-2
Inverter
Shift Reactor
High Temperature
Canal
Canal
Canal
Canal
Canal
Canal
Air Fin
S&T
S&T
Air Fin
Centrifugal
Centrifugal
Centrifugal
Packed Column
89.6 MJ/hr
7.0 MJ/hr
9.8 MJ/hr
31.7 MJ/hr
16.9 MJ/hr
290 MJ/hr
290 MJ/hr
158 MJ/hr
167 MJ/hr
167 MJ/hr
8.1 nm3/min
-------
CONVECTION SECTION
2 PASS
28 TUBES EACH
FLUE GAS
SHELL
0.32 cm C.S.
REFORMER
SECTION
(16 TUBES)
INSULATION
7.6 cm
FUEL FROM
ANODE
EXHAUST
REFORMATE
AIR
FIGURE IV-26. REFORMER PACKAGE FOR 100-kW POWER PLANT
IV-102
-------
combustion chamber. Vertical reformer tubes (7.6 cm o.d.) are uniformly
distributed within the combustion chamber at a center distance of
15.2 cm (6 in.). The chamber is lined with castable refractory.
Reformer catalyst (Girdler G-56-A) is the same as proposed for System 2
and its volume is based on a design space velocity of
3 3
1,600 m !!„ product/hr-m catalyst. The heat flux is
105 MJ/hr-m2 (9,250 Btu/hr-ft2), based on the tube i.d. of 5.9 cm
(2.34 in.).
A preheater for SNG and steam is located above the reformer
section. It is a two-pass cross counterflow design with the cooler SNG
mixture flowing through the 2.5 cm (1 in.) tubes and the hot flue gas
passing over the tubes.
That assembly is based on the following details of construc-
tion, from which materials costs are later estimated:
o Reformer tubes - 7.6 cm (3 in.) o.d., cast in HK 40
o Convection section - 2.5 cm (1 in.) tubes in 309 SS
o Insulation - 7.6 cm (3 in.) thick castable refractory
o Shell - 0.32 cm (0.125 in.) thick carbon steel
o Catalyst - Girdler 56A
The small zinc oxide guard bed was omitted from the conceptual design
and cost estimate.
Shift Reactors — The sizing of the high- and low-temperature
o
shift reactors is based on space velocities of 562 and 761 m inlet
o
gas/hr-m catalyst, respectively. These values were obtained from the
Girdler handbook using Type G-3A and G-66B catalysts. A length/diameter
ratio of four was used and the configuration is assumed to be a welded
carbon steel vessel with an internal insulating lining and baffle plates
to support the catalyst and distribute the gas.
IV-103
-------
Heat Exchangers — The heat exchanger dimensions were given in
Table IV-15. Heat Exchangers E-1A through E-4 are canal type designs,
which would be lighter and less costly than the shell and tube type.
Canal heat exchangers are roughly one-third as expensive as shell and
tube exchangers assuming the same surface area. They have lower pres-
sure capability and are generally used for heat recovery applications.
Materials shown in Table IV-15 assume a maximum temperature capability
of 343°C (650°F) for carbon steel. Heat transfer areas were based
on overall heat transfer coefficients of 200 kJ/hr-m -°C (10 Btu/
hr-ft2-°C) for gas/gas duty and 1,000 to 1,400 kJ/hr-m2-°C
(50 to 70 Btu/hr-ft2-°F) for liquid/liquid and air fin units.
System Packaging — A layout of the system was prepared assum-
ing that all components are assembled into a single unit. The estimated
package size, shown in Figure IV-27, is 2.6 x 3.7 x 3.0 m (8.5 x 12 x 10
ft). The package is assumed to have a welded steel base using channel
sections and a frame of steel angles forming a superstructure to support
heat exchangers, ducting, and enclosure. The package is designed for
outdoor installation. The size and weight of the package falls well
within the limits for transport by truck.
6. Heating and Cooling System
The 100-kW fuel-cell power plant described in the previous
section is designed to provide electricity and heat to an apartment,
condominium, or cluster housing complex. In our analysis, the resi-
dences are townhouses arranged in a cluster that minimizes the cost of
distributing heat recovered from the fuel cell.
The townhouses are similar in floor space and construction to
the detached houses described in Section IV-A, with the exception that
they share common walls. The effect of that arrangement on heating and
cooling requirements is to reduce the heat loss or gain as a function of
temperature relative to the detached houses. The heat loss parameter
IV-104
-------
f
h-'
o
Ul
FUEL CEtL STACKS
E5
B1
P.C.
-_«-_
FUEL
CELL
STACKS
[V
E1A
E2A
E4
FIGURE IV-27. SYSTEM LAYOUT FOR 100-kW POWER PLANT
-------
for the townhouses is 230 kJ/°C-hr (390 Btu/°F-hr), compared to
360 kJ/°C-hr (610 Btu/°F-hr) for the detached houses. The cor-
responding heat loss equation, which includes internal heat sources, is
Q = 23,900 - 230 T kJ/hr
(Q = 22,600 - 390 T Btu/hr),
where Q is the heat input required to maintain the house at 21 C
(70°F), and T is the external temperature in °C (°F).
Because the fuel-cell power plant is designed to allow heat to
be recovered, the most economical method of meeting the heating load is
to use recovered fuel-cell heat whenever it is available. When fuel-
cell heat is not sufficient, heat pumps are used to make up the differ-
ence. The heat pumps are also used to provide air conditioning in the
summer.
The arrangement chosen for this analysis is for 20 residences
to be served by a 100-kW fuel-cell power plant. The recovered heat is
distributed from a central location through a stream of 82°C (180°F)
water. In addition to space heating in the winter, the hot water stream
provides domestic hot water (DHW) throughout the year. The residences
are equipped with individual heat pumps. The site plan for the power
plant and townhouses is shown in Figure IV-28. The details of the hot
water distribution system are shown in Figure IV-29. As the hot water
is piped into a residence, it is distributed either to the heat ex-
changer in the DHW tank or to the heat exchanger in the central heating
duct. The flow is controlled by thermostats in the DHW tank and on one
of the interior walls of the residence. The availability of recovered
heat from the fuel cell is a function of the electrical load, which is
determined by the use of lights, appliances, and heat pumps in the resi-
dences.
Because of the lower heat gain or loss from the townhouses,
the heat pumps can be of smaller capacity than those described in
IV-106
-------
M
f
TOWNHOUSES
TOWNHOUSES
12.2m
J_
. J
POWER PLANT
T*
i
-1-
TOWNHOUSES
30.5m
—»~ HOT WATER DELIVERY
«•- - COLD WATER RETURN
TOWN HOUSES
FIGURE IV-28. SITE PLAN FOR 100-kW FUEL-CELL POWER PLANT AND TOWNHOUSES
-------
DHW
180°F
HEAT
PUMP
TO HOUSE
HEATING VENTS
AMBIENT
AIR
HOT WATER
FROM POWER
PLANT
COLD WATER
TO POWER
PLANT
FIGURE IV-29. SCHEMATIC OF HOT WATER AND SPACE HEAT SYSTEM
USING RECOVERED FUEL-CELL HEAT
IV-108
-------
Section IV-B. A heat pump that meets the design cooling load of 15.6
MJ/hr (14,800 Btu/hr) is one scaled to 3/4 capacity of the advanced
model described in Section IV-B. It has a rated cooling capacity of
19.3 MJ/hr (18,300 Btu/hr) at 35°C (95°F). Its heating and cooling
performance are shown in Figure IV-30, along with the heating and cool-
ing loads of the residences. The cooling load is based on average sum-
mer afternoon conditions of humidity and insolation (see Chapter VIII).
Two banks of 4.7 kW electric resistance heaters provide additional heat-
ing capacity below the balance point at -5°C (23°F). Their effect
is shown as dashed lines in Figure IV-30.
Other important components of the heat delivery system are the
heat exchangers that transfer heat from the 82°C fuel-cell hot water
stream to the DHW tank and the space heating system. The water-to-water
heat exchanger in the DHW tank was assumed to be a simple coiled copper
tube, which effectively transfers heat at the approach temperatures and
flow rates that characterize this system. The choice of a water-to-air
heat exchanger is more complicated, reflecting the many variables in the
system, including duct size, air flow rates, water flow rates, and
approach temperatures.
The water-to-air heat exchanger (or heating coil) was chosen
using available manufacturer's literature. It can effectively
transfer heat from hot water (82°C) with flow rates ranging from
0.0057 liter/sec (0.090 gal/min) to 0.070 liter/sec (1.11 gal/min), heat
pump exit air temperatures ranging from 21 C (70°F) to 43°C
(110°F), and a fixed air flow rate of 0.26 m3/sec (550 scf/min).
Furthermore, it was required to fit in a heating duct that was about
0.30 m (1 ft) square. The heating coil chosen for this application was
the Trane type WC, Series 18 (with turbulators), with 30 cm x 30 cm (12
in. x 12 in.) active area. The heat transfer characteristics of this
device are discussed in Chapters V and VIII.
Physically, the heating coil is located in the hot air duct
between the heat pump condenser coil and the supplementary resistance
IV-109
-------
EXTERNAL TEMPERATURE - °F
70 80 90 100 110
O 30 -
LU
Q
§ 20
I1
« ' 10
o
o
o
20 25 30 35 40
EXTERNAL TEMPERATURE -°C
30,000
20,000
€
a
10,000
45
-20
50
-10
EXTERNAL TEMPERATURE - °F
0 10 20 30
40
50
60
50,000
40
O
<
130
o
cr
o
CO
<
10
Oli-
SUPPLEMENTARY ELECTRIC
RESISTANCE HEAT
-30 -25 -20 -15 -10 -5 0
EXTERNAL TEMPERATURE - °C
10
40,000
30,000
20,000
10,000
15
FIGURE IV-30. HEATING AND COOLING CAPACITIES OF 19.3 MJ/hr
(18,300 Btu/hr) HEAT PUMP
IV-110
-------
heaters. Heating loads are met by either the heating coil alone, heat-
ing coil plus heat pump, or, on very cold days, heating coil, heat pump,
and resistance heaters combined.
IV-111
-------
F. References—Chapter IV
1. "Energy Alternatives: A Comparative Analysis," Science and Public
Policy Program, University of Oklahoma (1975).
2. G. R. Rowe, "Overall Economics of the Unit Train for Western Coal,"
Burlington Northern, Inc. (1975).
3. Proceedings of the 2nd International Technical Conference on Slurry
Transportation, Las Vegas, Nevada, March 2-4, 1977-
4. "Critique and Response to Coal Transportation," Peabody and
Associates, Inc., National Technical Information Service Number
PB-251-521 (April 1976).
5. "From Mine to Market by Rail...The Indispensible Transport Mode,"
Coal Age, p. 106 (July 1974).
6. A. J. Frabetti, "A Study to Develop Energy Estimates of Merit for
Selected Fuel Technologies," Development Sciences, Inc. (September
1975).
7. "Rail Transportation...Equipment," 1976 Keystone Coal Industry
Manual, p. 196.
8. W. H. Ponder, et al., "SO Control Technologies — Commercial
Availabilities and Economics," Third Annual Conference of Coal
Gasification and Liquefaction, Pittsburgh, Pennsylvania, August
1976.
9. "Coal-Fired Power Plant Capital Cost Estimates," Bechtel
Corporation, EPRI Report AF-342 (January 1977).
10. R. S. Jens, "System Planning: Transmission," Perspectives on the
Electric Utility Industry; A Handbook by Electric Power Research
Institute, p. 10-7 (May 1977).
IV-112
-------
11. M. L. Baughman, and D. J. Bottaro, "Electric Power Transmission and
Distribution Systems Costs and Their Allocation," IEEE Transactions
on Power Apparatus and Systems, p. 782 (May/June 1976).
12. Federal Power Commission, The 1970 National Power Survey, Part 1,
p. 1-13-9.
13. "Transmission Lines," from Encyclopedia of Energy, D. N. Lapedes,
ed., p. 698.
14. R. Detman, et al., "Factored Estimates for Western Coal Commercial
Concepts," Interim Report, Energy Research and Development
Administration (October 1976).
15. R. D. Howell, "Mechanical Design Consideration in Commercial Scale
Coal Gasification Plants," Sixth Synthetic Pipeline Gas Symposium.
16. W. L. Hecklen, "The Construction of Conversion Vessels," Energy
Communications, p. 133 (1976).
17. A. J. McNab, "Design and Material Requirements for Coal Conversion"
(November 1975).
18. R. N. Maddox, Gas and Liquid Sweetening (1974).
19. H. S. Kirschbaum, and S. E. Veyo, "An Investigation of Methods to
Improve Heat Pump Performance and Reliability in a Northern
Climate," Westinghouse Electric Corporation, EPRI Report EM-319
(January 1977).
20. J. M. King, Jr., "Advanced Technology Fuel Cell Program," EPRI
EM-335, Final Report (October 1976).
21. Girdler Catalysts, Chemtron Corporation, "Physical and
Thermodynamic Properties of Elements and Compounds," GC 245-10-869,
Rev. 3.
IV-113
-------
22. Girdler Catalysts, Girdler Chemical, Inc., "Hydrogen and Synthesis
Gas Production."
23. Stanford Research Institute, "Synthetic Petroleum for Department of
Defense Use," Defense Advanced Research Projects Agency Report
AFAPL TR-74-115 (November 1974).
24. Exxon Research and Engineering Company, "Evaluation of Methods to
Produce Aviation Turbine Fuels from Synthetic Crude Oils, Phases 1
and 2," Air Force Aero-Propulsion Laboratory/SSF WRight Patterson
Air Force Base, Ohio: Reports AFAPL-TR-75-10, Volumes 1 and II
(March 1975 and May 1976).
25. A. C. Antoine and J. P- Gallagher, "Synthesis and Analysis of Jet
Fuels from Shale Oil and Coal Syncrudes," U.S. Department of
Commerce, Report N76-21341 (August 1976).
26. E. R. Elzinga, et al., "Application of the Alsthon/Exxon Alkaline
Fuel Cell System to Utility Power Generation," Electric Power
Research Institute, Report EM-384 (January 1977).
27- United Technologies Corporation, "Advanced Technology Fuel Cell
Program," Electric Power Research Institute (October 1976).
28. J. G. Bendoraitis, et al., "Upgrading of Coal Liquids for Use as
Power Generation Fuels," Electric Power Research Institute Report
361-1 (January 1976).
29. D. T. Beecher, et al., "Energy Conversion Alternatives Study —
Combined Gas/Steam Turbine Plant Using Coal-Derived Liquid Fuel,"
NASA CR-134942 (November 1976).
30. J. Neal, "New Gas Turbines Could Provide Fuel Benefits," Public
Power, p. 28 (November-December 1976).
IV-114
-------
31. "Mixing NO ," Chemical Engineering, p. 70 (April 25, 1977).
X
32. EPRI Journal, p. 49 (August 1977).
33. Private communication, Pratt and Whitney Aircraft Division of
United Technology Corp.
34. Massachusetts Institute of Technology, "Economic and Technical
Aspects of Gas Turbine Power Stations in Total Energy
Applications," U.S. Army Facilities Engineering Support Agency,
FESA-RT-2013, p. 56 (January 30, 1976).
35. W. B. Wilson and W. J. Hefner, "Economic Selection of Plant Cycles
and Fuels for Gas Turbines," Combustion, p. 7 (April 1974).
36. S. Abens, et al., "Fuel Cell Stacks," Final Technical Report,
Contract No. DAAK02-74-C-0367 , Project No. 7763580, Energy Research
Corp.
37. L. B. Welsh, et al., "Optimization of Pt-Doped Kocite Electrodes
in H PO, Fuel Cells," Interim Technical Report, Contract No.
DAAG53-76-C-0014, p. 25 (January 1978).
38. Ibid., p. 2.
39. H. G. Petrow, et al., U.S. Patent No. 4082-699, April 4, 1978.
40. S. G. Abens, et al., "Fuel Cell Stacks," Semi-Annual Report,
December 1974-August 1975, ERC-7396-S, p. 22 (March 1976).
41. Ibid., Third Interim Report, ERC-7396-III, p. 5 (May 1976).
42. Ibid., Fourth Interim Report, ERC-7396-IV, p. 11 (February 1977).
IV-115
-------
43. H. C. Maru, et al., "Phosphoric Acid Fuel Cell Cathode," National
Fuel Cell Seminar Abstracts, San Francisco, California, July 11-13,
1978, p. 76.
44. B. S. Baker, Paper presented at Hybrid Vehicle Workshop, Los
Alamos, New Mexico, August 15, 1977.
45. J. C. Trocciola, et al., U.S. Patent 4000-006, December 28, 1976.
46. R. D. Breault, U.S. Patent 4017-663, April 12, 1977-
47. Ibid., U.S. Patent 4017-664, April 12, 1977.
48. United Technologies Corp., "Improvement of Fuel Cell Technology
Base," Technical Progress Report No. 2, FCR-0809, April 1, 1977 -
April 1, 1978.
49. S. Abens, et al., "High Temperature Molten Carbonate Fuel Cells":
Fourth Quarter Technical Progress Report-E-3-4 (March 1977); Fifth
Quarter Technical Progress Report-E-3-5 (July 1977).
50. H. C. Maru and B. S. Baker, "Status of ERC's Phosphoric Acid Fuel
Cell Technology," presented at Fuel Cell Workshop, Sarasota,
Florida, November 14-17- 1977.
51. Trane Corporation, "Application and Selection Data for Trane
Cooling and Heating Coils."
IV-116
-------
V. THERMAL EFFICIENCY OF THE SYSTEM COMPONENTS
Within each component of the five systems described in the previous
chapter, the energy-containing product is either moved from one place to
another or its physical and/or chemical form is altered. At each stage
of each system, some energy must be expended in carrying out these pro-
cesses. Ultimately, all energy entering the system as stored chemical
energy in the coal will be degraded to heat (random thermal motion) in
the environment. However, during this process of thermodynamic degrada-
tion (or loss in the capacity to do work), a certain amount of useful
work is obtained. In the systems under consideration in this study, the
desired end result is to provide heated or cooled air to the interiors
of residences. The effectiveness with which this goal is carried out
can be measured by the overall thermal efficiency of each system.
For purposes of this chapter, the thermal efficiency of an energy
transport or transformation process is simply the energy content of the
product of the process divided by the energy content of the product
entering the process, plus any additional energy contributed by another
source. This latter term would apply to diesel fuel consumed by a coal-
carrying unit train, for example. The five systems will be compared in
terms of their overall efficiency in providing residential heating and
cooling, starting with coal in the ground.
In subsequent calculations, the energy content of any energy-
containing product (solid, liquid or gaseous fuel) will be expressed in
terms of its higher heating valve (HHV), which is the amount of thermal
energy released upon combustion, including the heat of condensation of
water vapor produced during combustion.
The energy required to carry out any process that must be supplied
by an external source will be assessed only in terms of the direct fuel,
V-l
-------
heat or electricity consumed. No calculations will be made of indirect
energy requirements. Those are generally expressed as the energy em-
bodied in materials, chemicals, or human activities necessary for carry-
ing out the process.
Purchased fuel will be evaluated in terms of its higher heating
value, as previously discussed. Purchased electricity will be evaluated
in terms of the thermal energy required to produce it at the generating
plant, or approximately 10,550 kJ/kWh (10,000 Btu/kWh).
A. Coal Mine
The energy consumed in surface coal mining is in two principal
forms: (1) electricity for powering draglines, drills, and coal-loading
shovels, and (2) diesel fuel used in mobile equipment — coal-hauling
trucks, bulldozers, and scrapers. Minor amounts of energy are used in
heating and lighting shops and offices, powering small vehicles such as
foremen's trucks, and so on.
The use of electric power for the 4.5 million tonne (5 million ton)
per year mine described in Section IV-A is estimated to be 14 million kWh
per year. This represents an equivalent energy input of 1.5 x 10 GJ
Q
(140 x 10 Btu) per year.
Data from proposed western surface mining operations on consumption
of fuel by mobile equipment indicate that a 4.5 million tonne per year
surface mine would use about 3.8 million liters (1 million gallons) of
diesel fuel per year.2 Part of this fuel (5-10%) is used for
preparation of ammonium nitrate-fuel oil mixtures which are used as
explosive charges to loosen coal and overburden. Smaller equipment
consumes considerably less gasoline than diesel fuel — approximately
340,000 liters (90,000 gallons).2 The total energy content of these
two fuels is about 1.5 x 10 GJ.
V-2
-------
A 4.5 million tonne per year mine producing coal with a heating
value of 20.4 MJ/kg (8800 Btu/lb) produces the energy equivalent of
9.3 x 106 GJ (8.8 x 1012 Btu) per year. The "process" thermal ef-
ficiency is assumed to be 100%, because none of the coal is thermally
degraded during extraction.
The total consumption of liquid fuels and electricity during the
mining process is 3 x 10 GJ per year, or about 3.2% of the energy
output of the mine. The overall thermal efficiency, then, as defined at
the beginning of this chapter, is 97%.
B. Unit Train
The principal use of energy in shipping coal by unit train is the
diesel fuel consumed by the locomotives. To obtain a figure for the
quantity of fuel consumed it is customary to apply a factor of fuel con-
sumption per gross tonne-km (or ton-mi). The gross weight distance
factor includes the weight of the locomotives and cars for both the
delivery and return legs of the trip. Locomotives weigh approximately
140,000 kg (310,000 Ib) each and there are four per train. Coal cars
weigh about 28,000 kg (61,000 Ib) each. Thus, the gross weight of a
loaded 100-car unit train is 12,400 tonnes (13,700 tons), while the
weight of an empty train is 3,340 tonnes (3,670 tons). The total gross
weight distance for a 1,300-km (800-mi) shipment is thus 20.3 million
gross tonne-km (13.9 million gross ton-mi).
The appropriate figure for diesel fuel consumption for unit train
operation, assuming fuel-efficient throttling practices, is 3.6 liters
o
per 1000 gross tonne-km (1.4 gallons per 1,000 gross ton-miles).
Therefore, the round-trip fuel consumption for a 9,070-tonne (10,000-ton)
unit train carrying coal 1,300 km (800 mi) from the Powder River Basin
to the Kansas City-Omaha-Des Moines region is approximately 72,000
liters (19,000 gallons). This figure represents an energy consumption
of 230 kJ per tonne-km (320 Btu per ton-mi) of coal shipped, or a total
energy consumption equal to 1.5% of the heating value of the coal.
V-3
-------
Another use of energy in shipping coal is in the thawing sheds used
to warm coal in cars that have frozen during cold weather. Electrically
powered thawing sheds consume about 6.6 kWh per tonne (6 kWh per ton) of
coal and are typically capable of thawing five cars per hour. As a con-
servative assumption, thawing sheds are required on days when the aver-
age temperature is less than 0°C (32°F). Daily temperature data for
Omaha (see Chapter VIII) indicate that the average temperature is less
than 0°C (32°F) on about 85 days in an average year. Thus, a total
of (85/365) x 1.27 million = 0.30 million tonnes (0.33 million tons) of
coal per year will require thawing. The electricity consumption is
1.98 million kWh, or 20,900 GJ (20 x 109 Btu) thermal equivalent.
This figure represents 0.08% of the heating value of the coal shipped in
a year. This quantity is insignificant compared to the fuel consumed in
transporting the coal.
C. Coal-Fired Power Plant
New, uncontrolled fossil fuel power plants can have thermal effi-
ciencies approaching 40%. However, environmental control requirements
and the effects of age and cycling duty will reduce the actual thermal
efficiency well below this figure.
The net electrical output of the 800-MW coal-fired power plant de-
scribed in Section IV-C is estimated to be 34% of the energy represented
by the higher heating value of the coal input to the plant. The energy
balance for this facility is shown in Table V-l. The largest source of
thermal energy loss is in the steam cycle. Forty-five percent of the
energy input to the plant is lost from this source, mostly in the form
of heat ejected from the cooling towers. The next biggest loss, at 12%,
is boiler loss consisting of radiation losses and sensible and latent
heat in the flue gases. This figure is higher for subbituminous coal
than for bituminous coal because of the higher moisture content of sub-
bituminous coal, resulting in a greater latent heat content of the stack
gases.
V-4
-------
Table V-l
ENERGY BALANCE FOR AN 800-MW COAL-FIRED POWER PLANT
THAT USES SUBBITUMINOUS COAL
Energy Input GJ/hr Percent of Input
Coal 8,470 100
Energy Output
Product
Electric Power 2,880 34
Losses
Steam cycle losses 3,810 45
Boiler losses 1,020 12
Auxiliary power 420 5
FGD system 250 3
Electrostatic
precipitator 85 1
The use of electrical power in auxiliary equipment, such as boiler
feed pumps and coal pulverizers, results in a net loss of 5% of input
energy. The FGD system uses electrical energy equivalent to about 3% of
the total energy input. Finally, the electrostatic precipitator for
this facility, which must be large to handle the high-resistivity fly
ash associated with low-sulfur western coal, consumes power equivalent
to about 1% (maximum) of the energy input.
D. Coal Gasification Plant
The Hygas coal gasification facility described in Chapter IV is
designed to be self-sufficient in its energy needs, so no electricity is
purchased from an external source and fuel requirements are met by burn-
ing part of the coal brought into the plant or by-product oil. The com-
bustion of by-product oil derived from the gasification process as
boiler fuel reduces the amount of coal required for this purpose by
V-5
-------
about one-half. Thus, the overall plant efficiency based on the ratio
of SNG heating value to total coal heating value is increased. However,
plant economics are affected because the revenue that could be derived
from the sale of by-product oil is not realized. Environmentally, burn-
ing by-product oil is preferable to burning coal.
The overall energy balance for a 7.8 x 106 nm3 (275 x 106 scf)
per day plant is shown in Table V-2. The thermal efficiency for the
production of SNG from the coal entering the gasifier is 81%, reflecting
4
the high methane yield of the Hygas process. Because of the high
process efficiency, and the reduced requirement for coal as boiler fuel
due to the combustion of by-product oil, the overall plant thermal
efficiency is high — 74%.
Table V-2
ENERGY BALANCE FOR A 7.8 x 106 nm3 (275 x 106 scf)
PER DAY COAL GASIFICATION PLANT
BASED ON THE HYGAS PROCESS
Energy Input GJ/hr Percent of Input
Coal 14,900 100.
Energy Output
SNG 11,000 74.
Char 310 2.
By-products
(sulfur & ammonia) 80 0.5
Thermal Losses 3,510 23.5
Steam & Power Requirements 2,320 15.5
Coal 1,340 9.0
By-product oil 980 6.5
V-6
-------
Accounting for the plant's thermal losses is considerably more dif-
ficult than accounting for those in the coal-fired power plant. Heat is
released from numerous sources within the coal gasification plant; fur-
thermore, it is thermally integrated so that heat generated in one
portion of the plant is recovered and used elsewhere. For example, the
coal/water slurry feed to the gasifier is preheated using heat recovered
from the hot gases exiting the gasifier, and process steam is generated
using excess heat recovered from the methanation reactors.
An analysis of the various sources of heat release in the Hygas
plant indicates that, overall, about 20% of the thermal losses in the
plant are direct losses from boiler stacks, electric motors, sensible
heat of condensates and ash, and so on. The remaining 80% are indirect
losses in the form of wet or dry cooling, with the relative amounts of
these dependent on the location of the plant and the resulting cost and
availability of water. Using the integrated pollution control system
described in Chapter VI, in which a large fraction of the cooling tower
make-up water is derived from recycled process water, about two-thirds
of the plant's indirect heat loss is in the form of evaporative (wet)
cooling, and the remaining one-third is in the form of dry cooling. The
balance of wet and dry cooling will vary considerably, however, from one
plant design to the next.
The sources of heat within the plant that must be dissipated by wet
or dry cooling are many. Some of the main contributors are acid gas
removal, turbine condensers, and compressor interstage cooling.
E. Coal Liquefaction Plant
The conversion of coal to fuel oil via the H-Coal process is less
thermally efficient overall than the production of SNG via the Hygas
process. Although the conversion of coal entering the process stream to
end products has about the same thermal efficiency in both cases (around
80%), the steam, heat, and power requirements are about twice as high
for H-Coal as they are for Hygas for several reasons, including fuel
V-7
-------
needed for coal drying, higher compression energy requirements to attain
a coal slurry pressure of 18,600 kPa (2,700 psi) compared to 8,680 kPa
(1,260 psi) in the Hygas case, and the energy requirements for separate
gasification and steam reforming plants to supply hydrogen for the
liquefaction process.
The overall energy balance for a coal liquefaction plant that pro-
duces distillate fuel oil is shown in Table V-3. The overall thermal
efficiency of the plant is 66%. Of the 32.5% of the coal input energy
that is represented by thermal losses, 50-60% is accounted for by direct
heat losses (boiler stacks, coal drying, steam reformer stacks, electric
equipment, etc.) Of the remaining indirect heat loss, about one-half is
dissipated by wet cooling as shown in the integrated pollution control
system described in Chapter VI, and the other half is dissipated by dry
cooling.
Table V-3
ENERGY BALANCE FOR A 7950 m3 (50,000 bbl)
PER DAY COAL LIQUEFACTION PLANT THAT
PRODUCES DISTILLATE FUEL OIL
Energy Input GJ/hr Percent of Input
Coal 18,900 100
Energy Output
Fuel oil 12,400 66
Char 180 1
By-products
(sulfur and ammonia) 120 0.5
Thermal losses 6,140 32.5
Steam, Heat, & Power
Requirements 5,750 30
Coal 3,270 17
By-product gases 2,480 13
Hydrogen Requirements
Partial oxidation plant feed 4,200 22
(char and heavy bottoms)
Reformer feed
(by-product gases) 1,690 9
V-8
-------
Table V-3 also shows the energy requirements for steam, heat, and
power and for hydrogen production. A large part of the energy require-
ment (2,100 GJ/hr of by-product gases) is for fuel for the reformer that
drives the highly endothermic steam/hydrocarbon reforming reactions.
The energy balance for an H-Coal plant that produces hydrotreated
naphtha plus fuel oil is shown in Table V-4, whose data are similar to
Table V-3. The major differences are in the slightly higher steam, heat
and power requirements resulting from the addition of the naphtha hydro-
treater, and the larger feed of char and heavy bottoms to the partial
oxidation plant to provide additional hydrogen for hydrotreating. These
additional requirements reduce the overall thermal efficiency of coal
liquefaction to 64%.
Table V-4
ENERGY BALANCE FOR A 7630 m3 (48,000 bbl)
PER DAY COAL LIQUEFACTION PLANT THAT
PRODUCES NAPHTHA AND FUEL OIL
Energy Input
Coal
Energy Output
Naphtha
Fuel oil
Char
By-products
(sulfur and ammonia)
Thermal losses
Steam, Heat, & Power Requirements
Coal
By-product gases
Hydrogen Requirements
Partial oxidation plant feed
(char and heavy bottoms)
Reformer feed
(by-product gases)
GJ/hr Percent of Input
19,000 100
5,590 29
6,540 34
180 1
120 0.5
6,540 34
5,860 31
3,380 18
2,480 13
4,540 24
1,690 9
V-9
-------
F. Gas Pipeline
The 1300-km (800 mi) long, 81-cm (32-in.) diameter pipeline that
carries SNG from the Powder River Basin to the Omaha-Des Moines-Kansas
City region has 11 pumping stations. Between pumping stations the gas
pressure falls from 7,580 kPa (1,100 psia) to 4,100 kPa (600 psia).
Each pumping station requires a compressor capacity of 18.5 MW (24,800
o f.
horsepower) to recompress 943,000 nm (3.33 x 10 scf) per hour of
gas to 7,580 kPa (1,100 psia). Typically, the pumping stations are
powered by feeding part of the pipeline gas to the turbines that drive
the compressors.
The turbine-compressor units have a thermal efficiency of about
29%. Thus, 229 GJ (218 x 10 Btu) per hour of SNG must be consumed as
turbine fuel. For all 11 compressor stations, the total fuel require-
ment is 2,520 GJ (2,400 x 106 Btu) per hour, or 7.9% of the SNG that
flows through the pipeline. Therefore, the thermal efficiency of the
pipeline is 92%.
G. Liquids Pipeline
The operation of the 51-cm (20-in.) diameter fuel oil or naphtha/
fuel oil pipeline is similar to that of a gas pipeline, except that
pressure drops between pumping stations are greater — from 5,500 kPa
(800 psi) to 35 kPa (50 psi). However, the compression energy require-
ments for liquids are much less than for gases. The pumping station
capacity is only 2.4 MW (3,200 horsepower) for a mass flow rate 75%
higher than for the gas pipeline. Ten pumping stations are required for
the 1,300 km (800 mi) pipeline to transport 31,800 m3 (200,000 barrels)
per day of liquid fuels.
Typically, diesel fuel is purchased and supplied to the pumping
stations to run the diesel engines that power the centrifugal pumps. As
in the case of the gas pipeline, a diesel engine/centrifugal pump
V-10
-------
thermal efficiency of about 29% can be expected. The resulting diesel
fuel requirement is 788 liters (208 gal) or 29.6 GJ (28.1 x 106 Btu)
per hour. For the entire pipeline the fuel requirement is
296 GJ (281 x 106 Btu) per hour. This figure represents 0.59% of the
heating value of the fuel passing through the pipeline. The thermal
efficiency of the liquids pipeline is therefore 99.4%.
H. Liquid Fuel Distribution
The major source of energy use for the delivery of naphtha by truck
is the consumption of diesel fuel. Data on fuel consumption by gasoline
delivery trucks indicate that a 34,000-liter (9,000-gal) truck will
achieve an average fuel economy of 1.9-2.1 km/liter (4.4-4.9 mi/gal) in
a metropolitan area. Assuming an average round-trip distance of
80 km (50 mi) for the delivery of naphtha from a bulk storage terminal
to dispersed fuel-cell power plants (see Section VII-H), the diesel fuel
consumption per trip would be approximately 40 liters (11 gal) or
1.6 GJ (1.5 x 106 Btu). A 34,000-liter (9,000-gal) truckload of
naphtha has a heating value of 1,200 GJ (1,130 x 10 Btu). Therefore,
diesel fuel consumption is 0.13% of the heating value of the delivered
fuel.
Other sources of energy consumption and loss include pumping energy
and evaporation during loading and unloading. However, these quantities
are very small.
The delivery of distillate fuel oil from a bulk terminal to a com-
bined cycle power plant by unit train also requires consumption of
diesel fuel. Fuel consumption would be comparable to that of the coal
unit train discussed earlier, or 230 kJ per net tonne-km
(320 Btu per net ton-mi). Assuming a one-way distance of 160 km
(100 mi) (see Section V - H) and using a distillate fuel density of
0.85 kg/liter (7.1 Ib/gal), the diesel fuel consumption is
31 MJ/m^ (4,800 Btu/bbl) of fuel delivered. This figure represents
0.10% of the heating value of the distillate fuel.
V-ll
-------
I. Gas Distribution
No energy is consumed in the distribution of natural gas because
the gas enters the distribution system at pipeline pressure, which is
gradually reduced as the gas proceeds through the distribution system to
individual customers. No additional compression is required.
Some gas may be lost through leaks, but this amount is likely to be
small, and in any case would be very difficult to quantify. The effi-
ciency of gas distribution is assumed to be 100%.
J. Combined-Cycle Power Plant
The combined-cycle power plant achieves a high thermal efficiency
by using advanced, high-temperature gas turbines and by recovering a
substantial amount of the turbine exhaust heat in the form of high
pressure steam, which drives a steam turbine. The overall thermal
efficiency of the plant described in Chapter IV is 52% (heat rate =
6,920 kJ/kWh) at its rated load of 270 MW.7 The energy balance for
the plant is shown in Table V-5.
Table V-5
ENERGY BALANCE FOR A 270-MW
COMBINED-CYCLE POWER PLANT
Energy Input GJ/hr Percent of Input
Distillate fuel 1,870 100
Energy Output
Electric Power 979 52
Gas turbine generator 640 34
Steam turbine generator 330 18
Losses
Steam cycle losses 540 29
Stack gas losses 330 17.5
Auxiliary power 30 1.5
V-12
-------
The efficiency of the combined-cycle plant will vary with load.
Estimates of the heat rate at part-load operation have been made for a
similar combined-cycle plant that uses an advanced high temperature gas
Q
turbine. Those estimates have been used to derive values of the heat
rate at part loads (see Table V-6).
The main source of thermal loss in the plant is in the steam cycle,
which accounts for nearly a third of the energy input to the plant.
Most of this heat loss is dissipated in the cooling towers. The turbine
exhaust heat that is not recovered in the heat recovery steam generators
(17.5% of the energy input to the plant) is released with the stack
gases. The recovery of turbine exhaust heat is fairly high — 73%.
However, the subsequent steam cycle is relatively inefficient (39%), so
more than 60% of this heat is ultimately lost.
Even with several substantial sources of thermal loss, the
combined-cycle system is one of the most efficient electricity gener-
ating technologies. Although further improvements in turbine technology
are possible, efficiencies will probably not exceed 55%, due to
temperature limitations in the turbines and the inherent limitations of
the steam cycle.
Table V-6
ESTIMATED HEAT RATES AT PART LOAD FOR
COMBINED-CYCLE POWER PLANT
Percent of Rated Load Heat Rate (kJ/kWh)
100 6,920
75 7,020
50 7,040
25 7,650
V-13
-------
K. 26-MW Fuel-Cell Power Plant (SNG)
A final energy balance for this power plant is given in Table V-7.
Net power ouput is estimated to be 24.0-MW after allowances are made for
DC/AC inversion and transformation losses to 13.8 kV (AC) and auxiliary
power requirements. The system heat rate at full-load operation is es-
timated to be 7,920 kJ/kWh (7,509 Btu/kWh) using 930.6 kJ/g-mole CH4
(400,500 Btu/lb-mole CH ) as the higher heating value of the SNG feed.
Other process flow combinations might achieve the heat rate target,
but a substantial optimization effort would be required to define the
most cost-effective scheme. For example, lowering the cell voltage from
0.800 V to 0.787 V increases the current density to 150 mA/cm2. That
would reduce the total area of the fuel cell by 19%, thereby reducing
investment costs. However, the system heat rate would increase to about
8,010 kJ/kWh (7,595 Btu/kWh), exceeding the 7,910 kJ/kWh (7,500 Btu/kWh)
goal.
Table V-7
ENERGY BALANCE FOR A NOMINAL 26-MW
FUEL-CELL POWER PLANT THAT USES SNG
Energy Input GJ/hr Percent of Input
SNG 190 100
Energy Output
Electricity 86 45
Cathode exhaust 43 23
Air cooling 51 27
Inverter losses3 4.0 2
Auxiliary powerb 5.3 3
aTaken as 4% of DC power output, assuming inverter improvements.
^Primarily system blowers.
V-14
-------
The effect of part-load operation on the system heat rate was not
calculated. However, this variation has been estimated by United Tech-
nologies Corporation (UTC) for a similar molten carbonate fuel-cell
system.9 A minimum heat rate of 7,490 kJ/kWh (7,100 Btu/kWh) at 60%
of rated load was projected, as shown in Table V-8.
Lastly, the estimated values quoted above do not include efficiency
penalties reflecting possible fuel consumption required to maintain
system operating temperature during periods of stand-by operation. The
power plant load profile and a transient thermal balance are required to
evaluate this penalty.
L. 26-MW Fuel-Cell Power Plant (Naphtha)
The energy balance for this power plant is given in Table V-9. Net
power output is estimated to be 25.6 MW, with a system heat rate of
7,720 kJ/kWh (7,315 Btu/kWh). The effect of part load operation on the
power plant heat rate is similar to that shown in Table V-8.
The somewhat higher thermal efficiency of this power plant relative
to the SNG-fueled power plant is a result of the more favorable anode
reactant concentrations, as discussed in Chapter IV.
M. Electricity Transmission and Distribution
The main sources of energy loss between the generating plant and
the ultimate consumer are resistive and inductive losses in the high
voltage transmission lines and transformer losses in the distribution
system. The actual losses in a given utility system are a function of
many variables, including the transmission line voltages, distance from
generating plants to load centers, and average loads on transformers.
Rather than attempt to estimate such losses quantitatively, a more
reasonable approach is to use national statistics. On a nationwide
V-15
-------
Table V-8
UTC* PROJECTION OF PART LOAD HEAT RATE
FOR MOLTEN CARBONATE FUEL CELLS USING REFORMABLE FUELS
Percent of
Rated Load
100
75
60
50
25
Approximate
Heat Rate,
kJ/kWh (Btu/kWh)
7,910 (7,500)
7,540 (7,150)
7,490 (7,100)
7,560 (7,170)
8,280 (7,850)
United Technologies Corporation.
Table V-9
ENERGY BALANCE FOR A NOMINAL 26-MW
FUEL-CELL POWER PLANT THAT USES NAPHTHA
Energy Input
Naphtha
Energy Output
Electricity
Cathode exhaust
Air cooling
Inverter losses3
Auxiliary power
GJ/hr
198.0
92.0
44.0
52.0
4.1
6.1
Percent of Input
100
46
22
26
2
3
aTaken as 4% of DC output, assuming inverter improvements.
"Primarily system blowers.
V-16
-------
basis, the ratio of electricity metered to customers to that generatedby
utilities is about 0.91. This figure is commonly cited as the
average efficiency of electricity transmission and distribution, and
will be used in this study because more accurate methods of estimation
are not readily available.
The efficiency of distribution alone, appropriate where dispersed
fuel-cell power plants are used, is estimated to be 97%. Losses are due
mainly to energy dissipation in transformers.
N. 100-kW Fuel-Cell Power Plant with Heat Recovery
Energy balance calculations show that the 100-kW fuel-cell power
plant would generate 111 kW (DC) gross power to deliver 100 kW (AC) net
electrical power output at 120-220 V (AC). Corresponding electrical
energy conversion efficiency is 31.8%, based on the HHV of the SNG feed
consumed. At this rated load operating point, the waste heat recovery
system has a thermal output of 45.7 MJ/hr (443,000 Btu/hr), equivalent
to 4,915 kg/hr (10,827 Ib/hr) of hot water .product at 82°C. The cor-
responding thermal energy conversion efficiency is 40.3%. Thus, the
total input fuel energy utilization is 72.1% (HHV basis). The overall
energy balance for the power plant at rated load is shown in Table V-10.
The characteristics of the 100-kW power plant operating at part
load were assessed. Parasitic electrical energy losses were estimated
for the DC/AC inverter and power plant auxiliary equipment. Recovery of
waste heat thermal energy was also estimated. The results are given in
Table V-ll and Figure V-l. As shown, overall system energy utilization
exceeds 65% for power plant operation at 25-100% of rated electrical
load. Thermal output of the waste heat recovery system at part load
operation is shown in Figure V-2.
V-17
-------
Table V-10
ENERGY BALANCE FOR A 100-kW FUEL-CELL
POWER PLANT THAT USES SNG
Energy Input
SNG
Energy Output
Electricity
Recovered heat
Cathode/reformer exhaust
Inverter losses
Auxiliary power
GJ/hr
1.13
0.36
0.46
0.27
0.02
0.02
Percent of Input
100
32
40
24
2
2
Table V-ll
OPERATING CHARACTERISTICS OF 100-kW TOTAL
ENERGY POWER PLANT AT PART LOAD
Fuel Cell
Output, kW
(DC) Gross
111
83
56
28
11
Electrical Losses (kW)
Inverter
5.50
5.23
4.93
4.20
2.44
Auxiliaries
5.50
4.8
4.0
4.0
4.0
Power Plant
Output
kW (AC) Net
100
73
47.1
19.8
4.6
Input Fuel Energy
Utilization (%)
Electrical
31.8
32.5
32.4
28.4
17.3
Thermal
40.3
38.7
37.1
35.5
35.3
Total
72.1
71.2
69.5
63.9
52.6
V-18
-------
1—*
VO
100,
- 80
s
u
0>
a
I
60
z
UJ
o
UL
UJ
40
DC
UJ
X
20
82°C HOT WATER
(60°C WATER RETURN)
ELECTRICITY
I
I
20 40 60
NET ELECTRICAL POWER OUTPUT - kW (AC)
80
100
FIGURE V-1. EFFECT ON THERMAL EFFICIENCY OF OPERATING THE 100-kW TOTAL
ENERGY SYSTEM AT PART LOAD
-------
500 -
400 -
I
I-
01
I
Q
LU
CE
LU
O
O
III
CE
300 -
200 -
100 -
500
- 400
- 300
- 200
- 100
20
40 60
POWER OUTPUT - kW
80
100
FIGURE V-2. HEAT AVAILABILITY FOR 100-kW FUEL-CELL POWER PLANT
V-20
-------
Projected response of the 100-kW power plant to part-load operation
is similar to that reported by UTC for a 40-kW system. The UTC
system efficiency drops off more rapidly at very low load operation, so
that the estimated electrical losses given in Table V-ll may be optimis-
tic. A major difference exists in the estimated electrical conversion
efficiency. UTC reported a value of about 38%, but that was based on
the fuel's lower heating value. Conversion to an HHV basis would reduce
that figure to 34.2%. The remaining difference may be due to selection
of a higher cell voltage design point and/or extensive optimization of
the intergrated system components by UTC.
The impact of variable water return temperature was also assessed,
assuming that the thermal energy recovery heat exchanger design was
based on a 60 C (140 F) water return temperature. Increasing the
water return temperature (and keeping the exchanger area constant) had a
slight adverse effect on the overall system efficiency, as shown in
Figure V-3.
0. Gas Furnace and Air Conditioner
The rated efficiency of the gas furnace described in Section IV-A
is 80% -- 87 MJ/hr (82,500 Btu/hr) of gas is burned to yield 69.6 MJ/hr
(66,000 Btu/hr) of heated air. However, this is an instantaneous effi-
ciency achieved when the furnace has attained its appropriate operating
temperature. Because a furnace typically cycles on and off to meet its
heating load, it must start cold at the beginning of each cycle. During
start-up, while the plenum is being heated to its desired operating
temperature, no air is circulated, and most of the furnace heat is lost
through the flue. Over an entire heating season, these cold starts will
consume considerable additional energy. In addition, residential fur-
naces are not often maintained in optimum condition so that the rated
efficiency of the furnace is not usually attained over the life of the
furnace. Because of those and other smaller sources of energy losses,
the seasonal average energy efficiency for gas furnaces is estimated to
be 60%.
V-21
-------
to
ro
130
1001—
lao
a
I
u
§ 40
uj
I
£ 20
O
140
150
160
170
180°F
60
70
WATER RETURN TEMPERATURE
80° C
FIGURE V-3. EFFECT OF WATER RETURN TEMPERATURE ON OVERALL THERMAL EFFICIENCY
-------
Another source of energy consumption in a gas furnace is the blower
that is used to force air through the plenum and into the heating
ducts. The blower motor for the Westinghouse FGUE-082Q furnace is rated
at 895 kW (1/3 horsepower). Over the entire heating season, the
electricity consumption by the blower motor is about 7.8 kWh per GJ
(8.2 kWh per 10 Btu) of heat delivered by the furnace.
The energy efficiency of the combined Westinghouse SL030C condenser
and EC030 evaporator unit is taken directly from the manufacturer's
literature. The measure of this efficiency is the coefficient of
performance or COP- The COP is simply the ratio of heat removed from
the conditioned space to the heat equivalent of the electricity consumed
by the air conditioner. The COP provided by the manufacturer must be
adjusted, however, by including the electrical energy required to power
the fan that forces air through the EC030 evaporator unit, and by
reducing the cooling ouput by the heat generated by the fan motor. The
net effect is to decrease the rated capacity of the air conditioner from
30.9 MJ/hr (29,250 Btu/hr) at 35°C (95°F) exterior temperature to
29.8 MJ/hr (28,200 Btu/hr) and to increase the electrical load from 4.0
to 4.3 kW. The resulting COP for the SL030C/EC030 air conditioning
system as a function of exterior temperature is shown in Figure V-4.
Interior air temperatures are presumed constant at 26°C (78°F) dry
bulb and 19°C (67°F) wet bulb over the range of exterior tempera-
tures. The COP ranges from 2.25 at 27°C (80°F) to 1.58 at 43°C
(110°F). The COP decreases with increasing temperature because the
compressor must perform more work per unit of cooling as the temperature
difference between the higher temperature reservoir, to which heat is
exhausted, and the lower temperature reservoir, from which heat is
extracted, increases.
P. Heat Pumps
The advanced heat pumps described in Sections IV-B and IV-E have
excellent COPs, ranging from 1.7 to 3.6 as a function of external
V-23
-------
70
TEMPERATURE - °F
80 90 100
110
120
12
NJ
01
O
<
5
nc
o
u.
cc
01
a.
\-
LU
O
u.
LU
8
1 —
10
(C
LU
20
30 40
TEMPERATURE - °C
50
FIGURE V-4. COEFFICIENT OF PERFORMANCE OF WESTINGHOUSE
SL030 AIR CONDITIONER
-------
12
temperature. Current state-of-the-art heat pumps typically have
COPs in the range of 1.0 to 2.8. The COPs for the 26.0 MJ (24,600 Btu)
per hour and the 19.3 MJ (18,300 Btu) per hour heat pumps are shown in
Figure V-5 for both the heating and cooling modes. Because these heat
pumps have been been optimized for operation in a northern climate,
their performance in the heating mode is superior to that in the cooling
mode. The sudden drop-off in heating performance below 5.6°C (42°F)
is due to the effect of the defrost cycle, which must be engaged to
prevent frost build-up on the exterior evaporator coil.
As indicated in Section IV-B, it is not economical to size the heat
pumps large enough to meet all anticipated heating needs. Therefore,
supplemental heating must be provided to meet the heating load when it
exceeds the capacity of the heat pump. For the 26.0 MJ (24,600 Btu) per
hour heat pump that is used in the detached, single family homes de-
scribed in Section IV-B, that condition (called the "balance point")
occurs at about - 1°C (30°F). Below this temperature, resistance
heaters must be employed to balance the heat load. A bank of 3-4.7 kW
(16.9 GJ/hr or 16,000 Btu/hr) heaters would be capable of maintaining
the house at 21°C (70°F) at temperatures below -29°C (-20°F).
Because resistance heaters have a COP of 1.0, their use will lower
the effective COP of the heating system. This effect is shown in Figure
V-5 in which the dashed line shows the effective COP of the system below
the -1°F (20°F) balance point. The effective COP is an average over
time because resistance heaters are either on or off at any given
instant, cycling as required to maintain the interior temperature.
The balance point of the 19.3 MJ (18,300 Btu) per hour heat pump
that heats and cools the townhouses described in Section IV-E depends on
the amount of space heating supplied by recovered heat from the 100-kW
fuel-cell power plant. With no recovered heat, the balance point would
be at -5°C (23°F). For average light and appliance loads (1.01 kW)
and domestic hot water (DHW) demand (1.87 MJ/hr), as determined in
Chapter VIII, the balance point is shifted downward to -12°C (10°F).
V-25
-------
Below this temperature the supplemental resistance heater must be used.
The overall system COP is shown in Figure V-5, assuming no fuel-cell
heat is recovered, with the dashed line showing the effective COP below
the balance point of -5°C.
Naturally, the balance point of the system will vary with the
actual light, appliance, and DHW load. This variation is not great,
however, ranging from about - 13°C (9°F) to -11.5°C (11°F).
Q. Heat Delivery System
The key components of the system that delivers 82°C (180°F) hot
water from the fuel-cell power plant to the space heating and DHW
systems of the townhouses are the heat exchangers. The properties of
these heat exchangers determine the effectiveness with which the hot
water can be utilized.
The efficiency of the simple water-to-water heat exchanger used in
the DHW tank is assumed to be essentially 100%. This is a reasonable
assumption, because those heat exchangers are very efficient at the flow
rates and approach temperatures used in the system. As will be seen in
Chapter VIII, sufficient heat is available from the fuel-cell power
plant to meet all DHW demands during both the heating and cooling season.
The heat transfer efficiency of the water-to-air heat exchanger
described in Section IV-E is a function of the flow rate of the hot
water entering the exchanger and the temperature of the air flowing
across it. (It is also a function of the air flow rate and the water
temperature, but these quantities are constant in this system). The air
temperature is the same as the hot air stream delivered by the heat pump
and is a function of the external temperature. The hot water flow rate
is a function of the heat recovered from the fuel-cell power plant,
which in turn is determined by the electical load. The heat pump exit
air temperature varies from 26°C (79°F) at -29°C (-20°F)
external temperature to 43°C (110°F) at 16°C (60°F). The hot
water flow rate varies from 5.6 x 10 liter/sec (0.089 gal/min) at
V-26
-------
-20
20
TEMPERATURE - °F
40 60 80
100
UJ
o
cc
o
u.
cc
fe2
o
It
UJ
o
u
19.3 MJ/hr HEAT PUMP
WITH SUPPLEMENTAL
RESISTANCE HEAT
°30
-20
-20
-10
20
ui
u
<
I3
o
u.
C
8!
K
2
UJ
8
26.0 MJ/hr HEAT PUMP
WITH SUPPLEMENTAL
RESISTANCE HEAT
i
-30
-20
-10
10 20 30
TEMPERATURE - °C
40
120
10 20 30 40 50
TEMPERATURE - °C
TEMPERATURE - °F
40 60 80 100 120
14
12
10
8
6
4
2
0
14
12
10
8 I
50
FIGURE V-5. COEFFICIENTS OF PERFORMANCE OF ADVANCED HEAT PUMPS
V-27
-------
the minimum electrical load of 5.4 kW to 0.070 liter/sec (1.11 gal/min)
at 100 kW.
Using the heat transfer parameters supplied in the manufacturer's
literature for the Trane 30 x 30 cm (12 x 12 in.), WC Series 18 heating
coil (with turbulators), the water-to-air heat transfer rates were
calculated as a function of external temperature and electrical load.
The entering hot water temperature and air flow rates were held constant
at 83°C (180°F) and 0.26 nm3/sec (550 scf/min), respectively. The
results of the calculations are displayed in Figure V-6. The uppermost
curve in the figure is the heat available (per residence) from the
fuel-cell power plant and represents the upper limit on the heat that
can be transferred to space heating.
Figure V-6 displays several interesting features. The first is
that at a given electrical load (above 2 kW per residence) the
efficiency of heat transfer increases with decreasing external
temperature. That occurs because the temperature of air exiting the
heat pump decreases with decreasing external temperature, and the heat
transfer efficiency increases as the difference between air and water
streams. (The maximum heat transfer occurs when the heat pump is off,
since the air stream is near room temperature). This effect is
beneficial, allowing the most heat to be transferred precisely where it
is required most — at lower external temperatures, where the heating
load is greatest.
The heat transfer efficiency decreases with increasing heat
availability from the fuel-cell power plant because of the decline in
heat transfer effectiveness with increasing water flow rate. This
represents a disadvantage because the highest electrical loads occur at
low external temperatures where the heat demand is greatest.
Overall, the heat transfer efficiency ranges from a maximum of 100%
for various combinations of electrical loads and external temperatures,
to 49% at maximum electrical load and an external temperature of 16°C
(60°F).
V-28
-------
S3
25
20
15
UJ
u.
in
<
tr
10
HEAT
DELIVERED
HEAT AVAILABLE
I
I
234
ELECTRICAL LOAD PER RESIDENCE - kW
16(60)
25,000
20,000
15,000
3
m
10,000
5000
FIGURE V-6. HEAT TRANSFER RATES FROM HOT WATER DELIVERY SYSTEM TO SPACE HEATING
-------
R. References—Chapter V
1. Estimate of SRI International staff.
2. State of Montana, Department of State Lands, "Final EIS, Proposed
Plan of Mining and Reclamation, East Decker and North Extension
Mines, Decker Coal Co., Big Horn County, Montana," U.S. Department
of the Interior (June 1977).
3. M. E. Jacobs, "Fuel Efficiency Improvement in Rail Freight
Transportation: Multiple Unit Throttle Control to Conserve Fuel,"
NTIS PB 262-470.
4. R. Detman, et al., "Factored Estimates for Western Coal Commercial
Concepts," Interim Report, U.S. Energy Research and Development
Administration (October 1976).
5. R. F. Probstein and H. Gold, "Water in Synthetic Fuel Production —
The Technology and Alternatives" (The MIT Press, Cambridge, MA,
1978).
6. A. Melcher, et al., "Net Energy Analysis: An Energy Balance Study
of Fossil Fuel Resources," Colorado Energy Research Institute
(April 1976).
7. D. T. Beecher, et al., "Energy Conversion Alternatives Study —
Combined Gas/Steam Turbine Plant Using Coal-Derived Liquid Fuel,"
NASA-CR-134942 (November 1976).
8. Curtiss-Wright Corporation, "High Temperature Turbine Technology
Program — Phase I Topical Report," U.S. Energy Research and
Development Administration (January 1977).
9. J. M. King, "Advanced Technology Fuel-Cell Program," Electric Power
Research Institute Report EM-335 (October 1976).
10. Hittman Associates, Inc., "Environmental Impacts, Efficiency and
Costs of Energy Supply and End Use," Report No. HIT-561 (January
1975).
11. P. Bolan, "Heat Pumps and Fuel Cells," Paper No. 23d, AIChE 69th
National Meeting, November 30, 1976.
12. H. S. Kirschbaum and S. E. Veyo, "An Investigation of Methods to
Improve Heat Pump Performance and Reliability in a Northern
Climate," Electric Power Research Institute Report EM-319 (January
1977).
V-30
-------
VI ENVIRONMENTAL IMPACTS OF SYSTEM COMPONENTS
In this chapter, the environmental impacts of all the components
described in Chapter IV are analyzed. Wherever possible, the physical
factors (e.g., pollutant emissions) that are the source of environmental
problems are quantified. However, many of the problems addressed are
qualitatively discussed because quantification is difficult or impos-
sible. For systems components for which engineering controls are re-
quired to reduce environmental impact, those controls are described, and
their effect on the relevant environmental factors are quantified. All
control systems have been conceptually designed to meet or exceed cur-
rent and projected performance standards established by EPA.
The results of this chapter will be used in the comparison of the
five systems in Chapter IX. As discussed in Chapter I, EPA requested
that only physical environmental factors be used in the comparison of
the systems; therefore, we made no attempt to establish absolute levels
of impacts (e.g., human health effects).
A. Coal Mine
Several important environmental impacts result from strip mining
coal in the Powder River Basin, including:
o Air quality impacts
o Water resources impacts
o Land use impacts
o Topographic and geologic impacts.
These impacts are discussed in the following sections.
VI-1
-------
1. Environmental Setting
The headwaters of the Powder River, the major drainage for the
basin, begin in the Big Horn Mountains in north central Wyoming. The
river flows in a northerly direction, crossing into the state of Montana
and continuing until it drains into the Yellowstone River.
Bordered by the Big Horn Mountains to the west, the Black Hills to
the east, the Laramie Mountain Range to the south, and the Cedar Creek
anticline in Montana to the north, the Powder River Basin covers an area
of 36,066 km2 (13,914 mi2)in Montana and Wyoming.
The Powder River Basin lies in the Missouri Plateau physiographic
region. The terrain includes mountains, rough lands and badlands, hilly
areas, and moderately sloping lands. The average annual rainfall
25-41 cm (12-16 in.), comes mainly from thunderstorms that drop large
amounts of rain in short periods of time. The general climate of the
area, according to the Koppen climate classification system is middle
latitude steppes climate. Cut off by mountains from the invasion of
maritime air masses, the area is semiarid, with great annual temperature
variations between summer and winter. Its growing season is from 120 to
140 days long.
Water resources in the basin consist mainly of the Powder River.
The average discharge for the Powder River at Locate, Montana, from 1938
to 1979 was 17 m3/sec (600 ft3/sec) or 535 million m3 (434,700
acre-ft) per year. Ground water availability is greater than
31.5 liter/sec (500 gal/min) in the areas immediately adjacent to the
Powder River and falls off to less than 3.2 liter/sec (50 gal/min) for
most of the basin area.
The major land use is cattle and sheep ranching. The average ranch
size for this area is 2,950 hectares (7,280 acres). Small amounts of
hay are raised as winter feed for the local livestock.
VI-2
-------
Population distribution in the area is sparse with a density of 1.3
persons per square kilometer (3.4 per square mile). Major towns of the
basin include Gillette, Wyoming, population 7,194; Buffalo, Wyoming,
population 3,394; Midwest, Wyoming, population 825; and Broadus, Montana,
population 799.
2. Surface Mining Operation Requirements
The majority of the coal in the Powder River Basin lies close
enough to the surface that strip mining techniques are the most effec-
tive means of extracting the resources. Montana, Wyoming, and recently
the federal government have developed strict laws to regulate surface
mining. To narrow the scope of this analysis, it is assumed that the
coal resource will be mined in Campbell County, Wyoming, by an area
mining technology, and that the mine will supply 4.5 million tonnes (5
million tons) of coal per year to markets in the Midwest or to minemouth
conversion plants.
Mining companies are required to have a federal permit prior to
mining under the Surface Mining Control and Reclamation Act of 1977.
Companies applying for permits must meet the following conditions:
o Incorporation (e.g., partnership, corporation, individual)
o Established financial capabilities
o Mining rights
o A boundary map prepared
o Buyers for the coal
o Available equipment and personnel
o Coal transportation arranged.
The first step in the application process for a federal permit is
to design a field program for a baseline study to collect the data
VI-3
-------
necessary to meet federal requirements. The baseline studies should
examine the following areas:
o Hydraulic areas (surface and underground water)
o Subsurface geology
o Air quality
o Topography
o Archaeological and historical sites
o Socioeconomic impacts
o Land use impacts
o Noise
o Terrestrial biomes
o Aquatic biomes.
These areas can be addressed with the following procedures:
o Make literature searches
o Collect field data
o Write preliminary reports
o Devise a mining plan
o Prepare and submit a reclamation plan
o Assess the environmental impacts.
After these procedures have been completed, the following remains to be
done:
o Prepare and submit permit applications
o Advertise the applications
o Receive comments on the environmental impacts and the
reclamation plan.
The permitting agent then decides whether a hearing is necessary or
whether to grant the permit without a hearing. If a hearing is neces-
VI-4
-------
sary, it is advertised and held. The decision is then made on granting
of a permit. If the permit is denied, a hearing on the denial is adver-
tised and held. The permit process may take up to 2 years, or even
longer if litigation is involved.
3. Environmental Impacts of Surface Mining
Surface coal mining greatly affects the surrounding environ-
ment. Air-borne dust and particulates cause air pollution when large-
scale earth moving operations and wind erode the exposed and loosened
soil. Soil stripped of ground cover increases water runoff. Ground
water aquifers are disturbed by the excavation of overburden and removal
of the coal seams which often are aquifers. Land use is at least tem-
porarily converted from range and farmland to coal mining. Topographic
and geologic impacts occur after the removal of the coal itself. These
primary physical impacts are discussed below.
a. Air Resource Impacts
Several surface mining operations contribute to air pollution,
including emissions from coal-haul trucks and grading and drilling
equipment, as well as the dust created by excavation, grading, and con-
struction. Dust from the coal-haul roads and wind erosion of exposed
soil further increase air pollution. In addition, the electrically
powered shovels and draglines contribute considerable dust during earth
moving.
Table VI-1 shows the particulate emissions for existing mines and
projects emissions from mines that are planned to be in operation in
Campbell County, Wyoming in 1980 and in 1985.
The Wyodak Resources expansion for 1985 is projected to be mining
4.5 million tonnes (5 million tons) of coal per year. This concides
VI-5
-------
with the conceptual unit mining operation described in Chapter IV-A.
o
The particulate emissions are projected to be 375 x 10 kg/yr
(825 ton/yr).
The air quality is now considered to be good. The projected
expansion of current mining activity as indicated in Table VI-1 will
tend to considerably degrade air quality in the basin. Table VI-2 lists
air pollutant emission factors for major mining activities, assuming
uncontrolled emissions.
Table VI-1
PROJECTED PARTICULATE EMISSIONS FROM
SURFACE MINES - GILLETTE, WYOMING AREA
Source
Year; 1975
Amax Coal, South
Wyodak Resources
Year; 1980
Amax Coal, South
Wyodak Resources
Amax Coal, North
Carter Oil
Sun Oil
Kerr-McGee North
Year; 1985
Amax Coal, South
Wyodak Resources
Amax Coal, North
Carter Oil
Sun Oil
Kerr-McGee North
Comment
Existing
Existing
Expansion
Expansion
Expansion
Expansion
Expansion
Annual
Tonnes Mined
2,730,000
636,000
13,600,000
4,550,000
18,200,000
10,900.000
10,900,000
3,820,000
Particulate
Emissions
(103kg/yr)
225
52
Expansion
Expansion
New
New
New
New
9,270,000
2,270,000
13,600,000
7,270,000
9,090,000
3,820,000
760
190
1,130
600
750
315
1,130
375
1,500
900
900
315
Source: Reference 1
VI-6
-------
Wyoming's reclamation laws require that regrading and containing of
exposed soils and establishing of new ground cover begin as soon as pos-
sible. This plus the required treating of haul roads with chemicals and
watering will help control dust, although the dust created by the drag-
line remains a major contributor to air pollution (see Table VT-2).
b. Water Resources Impacts
Surface mining can cause major problems in surface and ground water
resources. The surface water runoff from a surface mining area is very
turbid. This problem is exacerbated by the pattern of rainfall in this
area, which typically occurs in the form of short, intense thunder-
storms. As much as 2 to 3 inches of rain can fall during a 24-hour
period, inundating the unprotected and exposed soils and carrying away
large amounts to local streams and then into the Powder River.
Ground water is affected more drastically in areas of shallow aqui-
fers, which are removed completely when they lie between the coal and
the surface. After the overburden is removed, the materials making up
the aquifers are mixed with the rest of the overburden. The ground
water recharge capacity is thus temporarily and possibly permanently
affected by this rearrangement of overburden material. Moreover, the
water quality may be affected by leaching of toxic substances that have
been disturbed in the mining processes.
Dewatering the mine can also result in the discharge of con-
taminated water, typically on the order of 30 liters/sec (500 gal/min).
Regulations proposed under the Surface Coal Mining and Reclamation Act
(30 CFR 715), limit the average concentrations of suspended solids,
iron, and manganese in such discharges to 30, 3.5 and 2.0 mg/liter re-
spectively. Because of the alkalinity of the water, however, the con-
centrations of iron and manganese are not likely to reach those limits.
Concentrations of 0.5 and 0.1 mg/liter have been estimated for iron and
o
manganese, respectively.
VI-7
-------
Table VI-2
AIR POLLUTANT EMISSIONS FOR A 4.5 MILLION TONNE
(5 MILLION TON) PER YEAR SURFACE MINE
Emissions (10^kg/yr
Activity
Dragline operation (annually
disturbing 100 acres at an
overburden depth of 70 ft)
Scrapers (3 in operation)
Haul road traffic
(haul trucks=1.0 kg/VMT)
(pickup trucks=0.4 kg/VMT)
Shovels and front-end loader
Trucks dumping at hopper
Vehicle exhaust
Haul trucks (assuming
6 45-ton trucks)
Pickup trucks
Scrapers, graders, and
loaders
Total vehicle exhaust emissions
Wind erosion of exposed soil
(0.24 tons/year-acre, assuming
800 acres exposed before
reclamation begins)
SOo NO,, Particulates Hydrocarbons CO
385
58
380
84
84
14
1
15
230
12
242
7.6
Negligible
0.79
8.4
13
.3
16
39
1
40
180
Note: VMT = vehicle miles traveled.
Source: References 1 and 2.
VI-8
-------
c. Land Use Impacts
Land in this area is primarily used for ranching and farming.
Native wildlife species including deer, antelope, and upland game birds
such as the sage hen make it their home range. The area is also a
stopover for migratory waterfowl. The reclamation process may take as
long as 4 years from the time the area is disturbed until it is regraded
and reclaimed.
The total area disturbed at any given time is about 270 hectares
(680 acres). Approximately 120 hectares (300 acres) of this area will
be semipermanent coal haul road, stage areas, and equipment maintenance
areas. Approximately 40 hectares (100 acres) will be newly mined each
year, and the other 110 hectares (280 acres) will be undergoing
reclamation.
Areas used for farming of particular crops such as wheat or hay may
not be restored to their original use. These areas may be restored to
grazing lands, but bringing them back to original crop production
capacity may be beyond the goals of the reclamation plan. The land may
be out of use during the mining process and also permanently decrease
productive crop acreage. In compensation, however, this action in-
creases range land for livestock and native wildlife.
When ponds are permanently eliminated by regrading the area, water
for livestock and wildlife is decreased. Waterfowl are displaced and
forced to seek a new habitat.
Oil and water wells are also in this region and may be temporarily
or permanently out of service if the land they occupy is mined.
d. Topographic and Geologic Impacts
Topographic and geologic impacts are related because the topography
of the land surface is changed in areas where thick coal seams are
VI-9
-------
extracted close to the land surface. The average overburden thickness
is 21 m (70 ft.). However, in some places it is only 6.1 m (20 ft.)
thick with an 18 m (60 ft.) coal seam under it. When the 18 meters of
coal are removed, a 24 m (80 ft.) hole with only 6.1 m of overburden
takes its place. Grading the adjacent areas to avoid the highwall com-
pensates for the remaining steep drop, but the resulting land surface is
lower. Disturbing an average of 40 hectares (100 acres) per year for 30
years means that an area of 1200 hectares (3,000 acres) of land is al-
tered. The handling of overburden during the mining process can involve
3 3
as much as 310 million m (339 million yd. ) of material (assuming
40 hectares (100 acres) per year at an average overburden thickness of
21m (70 ft.) for 30 years.
B. Unit Train (Coal)
Loading, line haul (transit), and unloading are activities asso-
ciated with movement of coal from the mine to the powerplant. The
severity of the environmental impacts of these activities is dependent
on the equipment involved as well as the location of the activity.
1. Loading/Unloading
The primary impact of loading and unloading is on air quality
because of the coal dust released. The dust generally affects only the
area immediately adjacent to the facility. This impact is not con-
sidered to be significant since the coal dust rapidly settles out of the
air.
2. Line Haul
Operating the unit train causes the most extensive
environmental impacts. Several different population groups as well as
VI-10
-------
the vegetation and wildlife along the route are affected. The groups
impacted include property owners along the right-of-way, residents in
nearby communities, and highway users. The new traffic intensifies
existing railroad-related impacts including noise and vibration, visual
intrusion, air pollution, and danger of accidents. The degree of impact
varies with the:
o Distance of affected persons from the railroad.
o Volume, scheduling, and type of railroad operations.
o Topography of the surrounding area.
o General condition of the right-of-way.
o Design and maintenance of the railroad's rolling stock.
o Level of maintenance of roadbed, track, and structures.
o Type of grade crossings.
o Land use of the surrounding area.
o Quality of construction and condition of buildings in adjacent
communities.
Railroad noise, together with the attendant vibration, often
annoys persons occupying nearby properties. Train horns and crossing
warning bells, the squeal of the train's brakes and of steel wheels
negotiating a curve, and switching operations all contribute to the
obtrusiveness of trains.
The noise profile around the railroad corridor varies with the
topography of the surrounding area, the weather, the degree of track
maintenance, the location of grade crossings equipped with warning
bells, the design of the railroad cars, and the roadbed. For example,
depressing the right-of-way relative to the surrounding land dampens
noise considerably, whereas elevating the track on a steel trestle
distributes the sound more widely. Deep setback of buildings from the
tracks, together with screening shrubs, lessens the perceived effect of
the train operations, although tests have shown that shrubs have little
VI-11
-------
actual effect on sound transmission. Also, a relatively high background
noise level tends to mask the train noises, so that railroad noises have
less effect in industrial and commercial districts.
Measurements of sound level made in Canada at a distance of
30m (100 ft.) from a freight train traveling at roughly the same grade
as the surrounding land are given in the following tabulation:
Sound Level
Source (dBA at 30 meters)
Train horn 98-100
Freight train—50 mph 90
Freight train engine—30 mph 87-92
Freight cars—30 mph 75-85
A noise level of about 90 dBA (the sound of train travel at 50 mph) can
cause workers to make significantly more errors than they would other-
wise. Noises above about 80 to 84 dBA are considered noticeable or
obtrusive. Outdoor noise levels of 70 dBA are considered to be a rea-
sonable maximum in residential neighborhoods by the U.S. Department of
Housing and Urban Development. These levels would obviously be sur-
passed when coal trains pass.
Danger involved with railroad operation almost exclusively
involves persons who cross the tracks, whether highway users, bicy-
clists, or pedestrians. This danger is of special concern to nearby
residents, employees, and customers because they are the people who are
most frequently exposed. Most pedestrian accidents happen to children
playing on the tracks or taking short cuts across them.
Other kinds of danger are associated with railroads, but are
usually not so important or as common as the hazard to pedestrians,
unless the community has experienced particular kinds of accidents such
as derailment. The effects are often more serious in rural areas where
speeds are higher than in urban areas.
VI-12
-------
A railroad line is quite visible and usually unattractive
unless depressed below ground level or buffered by buildings or
landscaping. The equipment is designed for functionality, not appear-
ance. Even when painted, rail cars seem like rolling billboards to
many. Dirt, rust, and lubricants frequently mar the right-of-way. The
motion of the train and its attendant noise attract attention.
Railroad rights-of-way are maintained for functional rather
than visual reasons. Poor weed control and wind-borne paper often add
to the usual litter of spilled lading and discarded railroad equipment.
This unsightliness often prompts local citizens to discard even more
waste in the right-of-way.
The smoke from the diesel locomotives visually intrudes in
local areas, both as it is emitted and as it blackens buildings and
structures. A well tuned and maintained diesel engine does not normally
emit smoke except under periods of heavy load, such as acceleration.
Thus, areas where the locomotives accelerate or where switching opera-
tions are conducted will be especially subjected to smoke. Switch
engines emit approximately 0.006 kg of particulate per km (0.02 Ib per
mi), and a fully loaded train emits about 0.14 kg per km (0.5 Ib per
mi). In comparison, a heavy diesel truck emits 0.0009 kg of partic-
ulates per km (0.003 Ib per mi).
Another source of air pollution from diesel engines is the
emission of unburned or partially burned hydrocarbon fuel. The average
emissions are 0.034 kg per km (0.12 Ib per mi) for switching service and
up to about 0.5 kg per km (1.8 Ib per mi) for fully loaded trains.
Comparable diesel truck emissions, 0.002 kg per km (0.007 Ib per mi),
are substantially greater per unit of load but are not so concentrated.
Again, the hydrocarbon emissions increase under acceleration or
hillclimbing, thus concentrating pollutants around these locations.
Other air pollutant emissions from diesel locomotives are shown in Table
VI-3.
VI-13
-------
As motor vehicles slow down or stop for grade crossings, then
accelerate back to speed, they emit more pollutants than they would were
they to continue along the same distance at a steady speed. These ef-
fects of the emission at grade crossings are felt in some measure all
over the air basin, but the effects are most pronounced near the sources
of the emissions. Therefore, this emission problem is both a neighbor-
hood and a community impact.
In addition to the pollutants discussed above, the unit coal
trains also add coal particles to air as they travel along the track.
Various estimates have been made of the amount of coal dust that is
lost, but there is little agreement on specific quantities because of
the number of variables involved. A figure of 1% of the coal load is
sometimes mentioned.
Table VI-3
AIR POLLUTANT EMISSIONS FROM
DIESEL LOCOMOTIVES—100-CAR UNIT COAL TRAIN
Emissions, kg/km (Ib/mi)
Pollutant Fully loaded Round trip average^
Particulates 0.13 (0.47) 0.085 (0.30)
S02 * 0.30 (1.1) 0.19 (0.68)
CO 0.68 (2.4) 0.43 (1.5)
Hydrocarbons 0.50 (1.8) 0.32 (1.1)
NOX 2.0 (6.9) 1.2 (4.4)
Aldehydes 0.029 (0.10) 0.018 (0.065)
*Assumes fuel sulfur content of 0.4%.
Source: Reference 2
VI-14
-------
C. Coal-Fired Power Plant
A coal-fired power plant can be a major source of air and water
pollution, and generates large amounts of solids that require dis-
posal. Table VI-4 lists the major emission sources for a coal-fired
plant that uses an FGD system to control sulfur emissions and a cooling
tower for steam condensation. The FGD system has been included even
though the plant is burning low-sulfur coal, because recent amendments
to the Clean Air Act indicate that, in the future, burning low-sulfur
coal will not be considered sufficient for sulfur emission control.
Figure VI-1 shows an integrated air and water pollution control
system that would be suitable for controlling most emissions. The
system does not include all of the possible technology; rather, it
reflects what could be readily installed in a new power plant. The
major feature of the treatment system is the use of the cooling tower,
boiler, and demineralizer blowdowns as make-up streams for the FGD
system and the boiler ash sluicing system. A chromium removal system
(reduction with SO and neutralzation to precipitate Cr(OH) ) is
included to treat cooling tower blowdown to limit plant discharge or to
protect the scrubber chemistry. That system minimizes fresh water in-
take and final effluent discharge and also minimizes the release of dis-
solved solids, because the FGD and boiler ash solids carry some blowdown
water to the solid disposal facilities as interstitial water.
Acidic runoff from the coal pile and the yard that contains dis-
solved iron, heavy metals, and sulfuric acid in addition to coal dust is
sent to ponds prior to neutralization and clarification. The treated
runoff is routed to the final effluent pond for recycling and discharge.
Miscellaneous chemical wastes from laboratory operations and condenser
tube cleanup are neutralized and released to the final effluent pond.
Two air emission sources — coal pile fugitive dust emissions and
dust from coal handling (which includes crushing for boiler injection)
— listed in Table VI-4 are not shown in Figure VI-1. Coal pile
VI-15
-------
Table VI-4
MAJOR SOURCES OF POLLUTION FOR A COAL-FIRED POWER PLANT
Medium
Air
Source
Emissions
Boiler Stack
Coal pile
Coal crushing and handling
Solid Boiler
Coal storage
Coal crushing and handling
Flue gas treatment
Water treatment
SOX
NOX
CO
Hydrocarbons
Particulates
Trace elements
Coal dust
Coal dust
Bottom ash
Coal fines
Heavy metal hydroxides
Coal fines
Fly ash
Desulfurization solids
Suspended solids
Water
Coal pile and yard runoff
Cooling tower blowdown
Boiler blowdown
Boiler feedwater treatment
Condenser tube cleaning
Flue gas treatment cleanup
Dissolved solids, heavy
metals
Suspended solids, organic
carbon compounds
Dissolved solids, suspended
solids, heavy metals
Dissolved solids
Dissolved solids
Dissolved solids
Suspended solids
Dissolved solids
VI-16
-------
EVAPORATION & DRIFT LOSS
EVAPORATION
t
CHROMIUM
CONTROL
DISCHARGE
130-190 LPS
MISC.
CHEMICAL
WASTES
•+- AMMONIA
-*-HYDRAZINE
.«- PHOSPHATE
DOMESTIC
WATER
SYSTEM
ASH SLURRY
5.7 LPS
MUNICIPAL
SYSTEM
SOLIDS
CONCEN
TRATION
SYSTEM
SULFURIC SODIUM
ACID HYDROXIDE
NEUTRALIZATION
SOLIDS TO DISPOSAL
OR CHEMICAL FIXATION
COAL PILE
RUNOFF
SOURCE: REFERENCE 4
LPS: LITERS PER SECOND
FIGURE VI-1. INTEGRATED POLLUTION CONTROL SYSTEM FOR AN 800-MW COAL-FIRED POWER PLANT
-------
emissions are uncontrolled, but emissions from handling are controlled
by using covered conveyers and processing facilities. Air in those
facilities is treated in a baghouse before release to the atmosphere.
The water requirements (and discharge) of the power plant shown in
Figure VI-1 are primarily related to the cooling tower. The cooling
tower shown has a concentrated factor of between 2 and 3 and therefore
has a significant blowdown. Depending on the chemistry of the actual
water supply, the concentration factor could be increased by 5 or 10
with either a simple lime-softening step on the makeup stream or by
withdrawing some recirculating cooling water for lime softening. Proper
choice of the concentration factor would allow the plant to discharge no
effluent to surface waters. All blowdown streams would be incorporated
as interstitial water in the solid waste stream.
The flue gas treatment system and the solids handling system, which
are the major pollution control systems on a power plant, are discussed
in detail below.
1. Flue Gas Treatment System
The flue gas contains approximately 80% of the ash content of
the coal in the form of a fine, particulate fly ash. Between 95 and
100% of the sulfur contained in the coal is in the flue gas in the form
of SO and SO.,. Other major components in the flue gas are NO , which
is created by high temperature reactions between atmospheric oxygen and
nitrogen in the fire box, C02, and water vapor. Some unburned hydro-
carbons and some CO are also present.
Emissions of CO and unburned hydrocarbons are controlled by
proper boiler design. No other control measures are practical at this
time. In new boilers, some control of NO emissions is possible by
X.
appropriate design. Flue gas recirculation to lower combustion temper-
atures, two-stage combustion that limits the oxygen concentration in the
hottest part of the flame, and careful control of excess air can reduce
VI-18
-------
NO emissions by 40-60%. These control techniques have been success-
fully put into practice on some oil and gas-fired boilers by retro-
fitting, and have been designed into coal-fired boilers built after 1971
when NO limitations were promulgated by the federal government. Ad-
X
ditional control of NO with those techniques, along with careful
X
burner design, may yield new coal-fired boilers with NO emissions
50-70-% lower than the current best practice. The additional NO
reduction probably will not be available to power plants now in exis-
tence, however. Further control of NO emissions is possible with
flue gas treatment — catalyzed and uncatalyzed reduction using ammonia,
and oxidation with ozone or hypochlorite processes have been partially
developed — but those systems are not likely to be installed on power
plants in the near future, because they are generally more expensive
than sulfur dioxide control systems. The power plant in Figure VI-1 is
assumed to meet currently mandated new source performance standards
(NSPS) of 0.30 kg NO /GJ (0.7 Ib NO /106 Btu) by a combination of
X X
good burner design, two-stage combustion and control of excess air.
The fly ash in the flue gas must be removed. For this pur-
pose, wet scrubbers, electrostatic precipitators, and baghouses are the
most suitable technologies. Wet scrubbers are not considered further
because collecting the fly ash in a dry form has several advantages, an
important one being that controlled addition of dry fly ash and certain
chemical additives to the sludge, which results from lime or limestone
scrubbing of flue gases (for removal of sulfur oxides), solidifies the
sludge and significantly aids in its disposal.
Baghouses have not been widely used in power plants because of
reliability and maintenance problems. The large space requirements of
baghouses have also limited their use. However, western coal, because
of its low sulfur content and the composition of its ash, produces a
high resistivity fly ash that is difficult to collect using standard
electrostatic precipitators. Accordingly, there is increased interest
in baghouses. Some plants designed to burn western coal are expected to
use precipitators (designed for the high-resistivity fly ash) and some
VI-19
-------
will use baghouses. The performance of the two is expected to be equi-
valent, and will readily surpass meet the current Federal NSPS for
particulates of 0.043 kg/GJ (0.1 lb/106 Btu) by removing 99% of the
fly ash from the flue gas.
There are many options for flue gas desulfurization. Lime or
limestone scrubbing to produce a solid calcium sulfite and sulfate phase
is the most developed technology, and the most likely to be installed in
the next 10 years. Double alkali systems and regenerative systems are
attractive alternatives and may eventually displace lime or limestone-
based systems. The double alkali system uses a soluble base to adsorb
S0_, which has at least two benefits: it avoids much of the pumping
and spraying of slurries, as well as scaling, that are associated with
lime/limestone systems, and the soluble base is more reactive than lime
or limestone, so lower recirculation rates and smaller scrubbers can be
used. The regenerative process produces a salable product such as sul-
fur or sulfuric acid rather than a solid waste (e.g., calcium sulfite or
gypsum) that has a limited market potential.
The lime/limestone scrubbing system can easily remove 85% of
the sulfur from the flue gases. The system in Figure VI-1 is based on
lime, which is more reactive than limestone. The scrubber removes pri-
marily SO and SO-, but also takes out some flue gas particulates that
were not removed by the electrostatic precipitator or baghouse. We have
assumed that it removes 50% of the remaining fly ash, which means that
the scrubbing slurry is approximately 1% fly ash.
2. Solids Handling System
The recirculating slurry used in the FGD unit is 5-10% solids.
A slurry blowdown is required to remove the CaSO- and CaSO, result-
ing from reaction of S02 with the added lime. Calcium sulfite is the
primary reaction product, but some oxidation of the sulfite to sulfate
occurs. Oxidation rates are increased by high 0 /S09 ratios, so
VI-20
-------
western coal, with its low sulfur content, produces a slurry with higher
CaSO./CaSO., ratios than the slurry from the desulfurization of high-
sulfur coals. The higher calcium sulfate content aids in dewatering the
slurry, which facilitates its disposal.
For the system shown in Figure VI-2, we have assumed that 40%
of the CaSCL is oxidized to CaSO,. The slurry can be thickened and
dewatered to a 50% solids cake on a rotating vacuum filter, after which
it can be mixed with the dewatered bottom ash slurry and the dry fly ash
to produce a waste stream containing only 31% water. This solid waste
can be handled with front-end loaders and trucks for disposal in the
landfill or transported back to the mine in coal cars for disposal in
the mine. Care must be exercised in the disposal, because the calcium
salts, the heavy metals from the fly ash, and the soluble salts in the
cooling tower blowdown can be leached from the material if it contacts
groundwaters or surface waters.
Other disposal options include placing the bottom ash and
scrubber blowdown slurries in ponds. Settling occurs in the ponds and
the clear supernatants can be recycled. The material in the pond bottom
is about 50% water. When the ponds are full they can be abandoned as
active disposal sites or dredged (the dredged solids are either disposed
by landfill or returned to the coal mine). However, abandonment results
in land that probably can never be returned to productive use.
Either ponding option is environmentally more risky than the dry
handling option because of the potential for pond water to overflow
during severe rainstorms or leach through the pond bottom into the
groundwater. Currently, no regulations prohibit ponding, but the EPA
considers it to be only a temporary solution.
3. Pollutant Emissions
a. Major Pollutants
The major emissions from a coal-fired power plant using the
pollution control methods discussed in the preceding sections are shown
VI-21
-------
RECYCLE
BOTTOM ASH
SLURRY
5000 kg/hr ASH
5000 kg/hr WATER
RECYCLE
SCRUBBER
SLOWDOWN
SLURRY
11,400 kg/hr SOLIDS
11,400 kg/hr WATER
VACUUM FILTER
DRY FLY ASH
19,800 kg/hr
MIXER
36,200 kg/hr SOLIDS
16,400 kg/hr WATER
FIGURE VI-2. SOLIDS HANDLING SYSTEM FOR 800 MW COAL-FIRED POWER PLANT
VI-22
-------
TABLE VI-5
SUMMARY OF MAJOR EMISSIONS FROM AN 800-MW COAL-FIRED POWER PLANT
Source
Steam Boiler
S02 - 803
NOX
CO
hydrocarbons
fly ash
Coal Handling
coal dust
Coal Storage
coal dust
Solids Disposal
dry FGD solids
dry fly ash
dry boiler ash
interstitial
water
Total
Water Discharge3,b
suspended solids
oil and grease
iron
copper
chlorinec
Uncontrolled
Emissions
(kg/hr)
5,200
19,900
160
70
11,400
19,800
4,980
16,400
52,580
Control Method
FGD unit
burner design
precipitator
plus FGD
bag house
none
burial in
controlled
landfill or
mine
Control Controlled
Efficiency Emissions
(%) (kg/hr)
neutralization
and
settling
85
99.5
99
780
2,560
207
62
100
1.6
70
11,400
19,800
4,980
16,400
52,580
20
10
0.7
0.7
0.3d
aRecycle pond overflow when cooling tower is operated at low cycles of concentration
Concentrations are based on BPCT standards in Reference 5.
bfiased on 190 liter per second (300 gal/minute).
c.2 ppm for a maximum of 2 hr/day.
dkg/day.
VI-23
-------
in Table VI-5. We calculated the emissions of SO and fly ash from
the known concentrations of sulfur and ash in the coal. The emission
of NO assumes a burner that meets the Federal NSPS for NO of 0.30
x , x
kg per GJ (0.7 Ib per 10 Btu) of fuel combusted. Emissions of CO and
hydrocarbons are based on standard emission factors for coal-fired
2
boilers. Fly ash and S0_ emissions are significantly lower than
the current Federal NSPS 0.043 kg/GJ (0.10 lb/106 Btu) and 0.52 kg/GJ
(1.2 lb/10 Btu), respectively. The control methods for those pol-
lutants are designed to meet not only current but also anticipated
future emissions standards for coal-fired boilers*.
In addition to the emission pollutants listed in Table VI-5, there
are emissions of several types of pollutants not currently subject to
regulation but known to have harmful environmental and human health
effects. Among these are fine particulates, toxic trace elements and
polycyclic aromatic hydrocarbons (PAH), which will be addressed in the
following sections.
b. Fine Particulates
The emission of fly ash from coal combustion involves the
mobilization of suspended particulate matter (or particulates) ranging
in size from greater than 30 microns in diameter to less than 1 micron
into the stack gas. However, the removal efficiency of fly ash collec-
tion equipment such as the electrostatic precipator tends to decrease
with decreasing particle size so that the smaller particulates are
preferentially emitted. Small particulates are of concern because they
are retained in the lungs. The range of particulate sizes that have the
highest lung retention are from about 0.1 to 5 microns in diameter, with
*In late 1978, the EPA issued a proposed revised set of NSPS that would
require FGD for nearly all coal burned, regardless of sulfur content and
would limit NOX emission to 0.26 kg/GJ (0.6 lb/106 Btu) and partic-
ulate emissions to 0.013 kg/GJ (0.03 lb/106 Btu).
VI-24
-------
peak retention occurring at around 1 to 2 microns. Particles larger
than 5 microns are not retained to any significant degree, and therefore
are of less concern as a health hazard.
Measurements of particle size distribution in fly ash from coal
combustion have shown that the mass-weighted median diameter for par-
ticulates entering an electrostatic precipitator is around 20 microns,
while for those exiting the median diameter is 1 to 5 microns. Further-
more, of the entering particulates, only 4 to 10% (by mass) are less
than 5 microns in diameter, while about 50% of the exiting particulates
are less than 5 microns in diameter. Thus, although electrostatic
precipitators may have removal efficiencies for total particulates in
excess of 99%, the efficiency for particulates less than 5 microns in
diameter is 90 to 95%. Applying those results to the emissions from the
800 MW coal-fired power plant, and adjusting them to the assumed col-
lection efficiency for that plant, yields an emission rate of parti-
culates less than 5 microns diameter of approximately 50 kg/hr.
c. Trace Elements
The emission of trace elements from a coal-fired power plant
depends on the concentrations of those elements in the coal as well as
on the various physical-chemical processes taking place during combus-
tion and fly ash collection. Certain trace elements tend to be con-
centrated on fly ash relative to bottom ash, and are concentrated
further on the fly ash exiting the electrostatic precipitator. Also,
some of the more volatile elements are emitted from combustion as gases
and escape the precipitator. Consequently, the emission rate of certain
trace elements is higher than one would calculate based on a uniform
distribution of trace elements in bottom ash, collected fly ash, and
emitted fly ash.
The concentration of several trace elements in Wyoming sub-
bituminous coal is shown in Table VI-6. Although many more trace ele-
ments are found in coal, those in Table VI-6 are listed because of their
VI-25
-------
known toxic effects. The concentrations of trace elements in coal
varies widely, and the concentrations in Table VI-6 should be considered
representative rather than definitive.
To estimate the rate at which the trace elements listed in
Table IV-6 are emitted in the power plant stack gases, one must know the
degree to which they are concentrated in the fly ash that escapes the
electrostatic precipitator. Of several studies that have been made of
the phenomenon, the one that was made under conditions most closely
matching the conceptual power plant design of this study was of trace
element distribution in a 350 MW power plant burning Wyoming sub-
bituminous coal and using an electrostatic precipitator (99.1% effici-
o
ency) for fly ash emission control. That plant's enhancement factors
(concentration in emitted fly ash relative to concentration in coal ash)
for the trace elements listed in Table VI-6 were as follows: Sb, 5.6;
As, 0.071; Be, 2; Cd, 5; Pb, 11; Hg, 140; Se, 54; Zn, 3.7. We assumed
that the enhancement factors apply to the electrostatic precipitator
only. No further enhancement is assumed for the FGD system, which is
assumed to remove 50% of the fly ash exiting the precipitator. In
general, trace element emissions depend on the fly ash removal system
employed (e.g., baghouses, Venturi scrubbers) as well as the specific
properties of the coal and ash.
Table VI-6
CONCENTRATIONS OF SEVERAL TOXIC TRACE ELEMENTS IN
POWDER RIVER BASIN COAL
Element Concentration (ppm wt.)
Antimony (Sb) 0.67
Arsenic (As) 3.0
Beryllium (Be) 0.7
Cadmium (Cd) 2.1
Lead (Pb) 7.2
Mercury (Hg) Q.l
Selenium (Se) Q.73
Zinc (Zn) 33.0
Source: References 7 and 8.
VI-26
-------
We used the concentration factors plus the total rate of
particulate emissions from the power plant to calculate the emission
rate of toxic trace elements. The total rate of emission of those ele-
ments in the stack gas as well as in the solid waste stream (bottom ash
plus collected fly ash) is shown in Table VI-7.
Another source of potential trace element pollution is leach-
ing from the solid waste disposal site. The significance of this
problem will depend on the permeability of the soil on site, the amount
of annual precipitation, the chemical forms of the trace elements and
their resulting solubilities, and the distance to the nearest aquifier.
Studies on the leachability of trace elements from coal ash and scrubber
sludge have indicated that, of the trace elements listed in Table VI-6,
Q
mercury and selenium are the most likely to pose problems.
Regulations proposed under the Resource Conservation and
Recovery Act of 1976 for the disposal of hazardous wastes may verylikely
be applied to coal combustion wastes. In that case, strict disposal
practices would be required.
Table VI-7
EMISSION OF TOXIC TRACE ELEMENTS FROM
AN 800 MW COAL-FIRED POWER PLANT
Emission Rate (g/hr)
Element
Sb
As
Be
Cd
Pb
Hg*
Se
Zn
Solid Waste
271
1,240
287
852
2,850
1
282
13,450
Stack Gas
6.2
0.35
2.3
17.0
130.0
40.0
20.0
200.0
*Hg is emitted primarily as a gas. The FGD system is assumed to have no
effect.
VI-27
-------
d. Polycyclic Aromatic Hydrocarbons
The final minor emission of concern from coal combustion is of
a class of compounds known as polycyclic aromatic hydrocarbons (PAH).
This class includes both known (e.g. benzo(a)pyrene) and suspected car-
cinogens. The emission of PAH results from incomplete combustion and
varies depending on burner type, amount of excess air used, and other
process variables. Measurements of the emission of several types of PAH
from a coal-fired power plant using a standard front-wall burner and
equipped with an electrostatic precipitator have been made. Those
results, scaled to the 800 MW size assumed for this study, are shown in
Table VI-8.
D. Coal Gasification Plant
A coal gasification plant has the potential for producing more
pollution than a power plant that consumes an equivalent amount of
coal. There are two major sources of additional pollution: dissolved
and suspended solids in process condensates, and organic and sulfur
emissions that result from clean up of the synthesis gas before
methanation.
Table VI-8
EMISSIONS OF POLYCYCLIC AROMATIC HYDROCARBONS FROM
AN 800-MW COAL-FIRED POWER PLANT
Compound Emission Rate, g/hr
Fluoranthrene 0.61
Pyrene 1.4
Benzo(a)pyrene 0.14
Benzo(e)pyrene 0.18
Benzo(ghi)pyrene 0.053
Source: Reference 10
VI-28
-------
The production of SNG requires contact of the coal with steam and
oxygen at high temperature and pressure. Some of the steam reacts with
the coal, but an excess is required to establish favorable reaction
conditions. Not all coal, steam, and oxygen reactions yield CH,, CO
and H . Tar, oil, light hydrocarbons, aromatics, polycyclic aromatics,
cyanide, ammonia, carbonyl sulfide, and hydrogen sulfide are also pro-
duced. All those compounds are partially soluble in water. When the
gases leaving the gasification reactor are cooled to remove oil and en-
trained ash, the water and soluble organics and inorganics condense.
The soluble materials and the ash must be removed from the water before
it can be reused or released to the environment.
The clean-up of the synthesis gas has the potential for releasing
hydrocarbons and reduced sulfur compounds such as H S and COS to the
air. The release of sulfur, if uncontrolled, would be the same as would
occur from simply burning the coal, but the reduced sulfur compounds are
more objectionable because they are both toxic and malodorous. The hy-
drocarbon emissions are caused both by losses of solvent in the synthe-
sis gas clean-up step (usually Rectisol or Selexol), and by the hydro-
carbon levels in the synthesis gas, which are higher than they would be
in flue gas from coal combustion. Some of those hydrocarbons are re-
moved by the gas clean-up solvent and released into process vent streams.
Table VI-9 is a detailed listing of the major sources of air,
water, and solid pollution associated with SNG production. Figure VI-3
is a schematic diagram of an integrated system for limiting water pol-
lution, primarily by eliminating discharge to surface waters, minimizing
air pollution, and disposing of solid residuals in a form that could be
placed in the mined-out area of the coal mine. Proper design choices
for the system could limit water discharges to reuse applications and
water in the solids stream.
Such a pollution control system contains many features common to
the system proposed for the coal-fired power plant. In particular, the
cooling tower is used to evaporate excess water and much of the water
blowdown from boilers, and ash handling and FGD residuals are contained
VI-29
-------
Table VI-9
MAJOR SOURCES OF POLLUTION FOR A COAL GASIFICATION PLANT
Medium
Air
Source
Emissions
Solid
Water
Boiler and heater stacks
SNG Process train
(including sulfur recovery plant)
Coal handling and crushing
Gasifier
Boilers and heaters
Flue gas treatment
Sulfur recovery plant
Coal storage
Wastewater treatment system
Cooling tower blowdown
Water treatment blowdown
Boiler blowdown
Process condensates
Boiler ash handling
FGD system
Coal pile
SOX
NOX
CO
Hydrocarbons
Particulates (fly ash)
Trace elements
S02
H2S
COS
Hydrocarbons
Coal dust
Bottom ash
Char
Bottom ash
Fly ash
FGD solids
Sulfur
Coal fines
Heavy metal hydroxides
Biological solids
Heavy metals
High molecular weight
organic compounds
Dissolved solids
Suspended solids
Heavy metals
H2S
HCN
Thiocyanate
Organic carbon
compounds
Suspended solids
Suspended solids
Dissolved solids
Heavy metals
Dissolved solids
Suspended solids
Dissolved solids
(heavy metals)
Suspended solids
Organic carbon
compounds
VI-30
-------
EVAPORATION 190 LPS
I
u>
RAW WATER
BOILER SLOWDOWN
I 28 LPS
COOLING TOWER MAKEUP
COOLING TOWER
SLOWDOWN
17 LPS
WATER TREATMENT SLOWDOWN '
PROCESS CONDENSATES 190 LPS
LIQUID PRODUCT
SCRUBBER MAKEUP 16 LPS
3.8 LPS EVAPORATION
| /I STACK
. " <—I RECOVERED WATER
FGD SLUDGE
ASH
SLUICE
MAKEUP
BIOLOGICAL SLUDGE \
CONCENTRATION
SOLIDS TO
13.1 LPS DISPOSAL
LPS: LITERS PER SECOND
COAL PILE FINES
FIGURE VI-3. INTEGRATED AIR, WATER, AND SOLIDS POLLUTION CONTROL SYSTEM FOR A HYGAS COAL GASIFICATION
PLANT THAT PRODUCES 7.8 MILLION nm3 OF SNG PER DAY
-------
in the solids stream that is disposed in the coal mine. The coal gasif-
ication plant has a utility plant that burns coal and liquid hydro-
carbons to generate power, steam, and process heat. The flue gases from
those combustion operations are treated to remove particulates and sul-
fur in the same manner as for the power plant in Section VI-C. Coal
pile runoff and particulate emissions from the coal pile and grinding
and drying are handled in the same way as they were for the power plant.
The two major additions to the pollution control system are a
sulfur recovery plant and its associated tail gas treatment system,
which treats sulfur-containing gases produced by synthesis gas cleanup
and by the process condensate treatment system.
1. Sulfur Recovery Plant
The raw synthesis gas produced by gasification contains
reduced sulfur compounds, primarily H S, and an excess of C09 in
addition to the required H and CO. All sulfur compounds must be
removed before the synthesis gas can be converted to methane. Numerous
processes have been proposed to remove the CO and H S, which together
are called acid gases. The processes include scrubbing with molten car-
bonates, scrubbing with amine solutions such as diethanolamine (DEA),
and scrubbing with organic solvents such as methanol. The process
proposed for all commercial-sized gasification plants is the Rectisol
process which uses refrigerated methanol. The Rectisol process makes
use of the different solubilities of H S and CO in methanol to
separate the acid gases into three product streams (in addition to the
purified synthesis gas) — an l^S-free stream which contains CO and
hydrocarbons, an H2S-rich stream which contains most of the H_S
(10-15% by volume) as well as some C02 and methanol, and an H2S-lean
stream (1-2% by volume), that contains much of the CO and some hydro-
carbons. The process can be operated to concentrate more than half of
the C02 in the stream that is free of H~S.
VI-32
-------
Downstream treatment of those three streams to remove the
sulfur involves a choice of two commercially proven processes, Glaus and
Stretford. The Glaus process can remove 95% of the H S in a stream,
along with much of the COS. However, it requires minimum input concen-
trations of 10-15% H S. The H S-rich stream produced by the
Rectisol process could be treated by a Glaus process, which removes some
COS and most H S (about 90% at the low concentration in the H S-rich
stream). The Stretford process is more versatile and could treat the
H S-lean stream; however, it is difficult to operate, materials such
as the cyanate and thiosulfate are formed by chemical reactions in the
Stretford solution and require a purge and make-up to maintain the
solution activity, and it will not remove COS. Neither process removes
hydrocarbons. A treatment sequence to achieve low levels of sulfur
emission as well as control of hydrocarbons would involve Glaus
treatment of the H S-rich stream (because that stream contains most of
the COS). The Stretford process would be used to treat the H S-lean
stream, followed by incineration of both tail gases. The H S-free
stream would be vented to the atmosphere. Incineration destroys the
hydrocarbons and converts reduced sulfur compounds such as H S and COS
to SO^. Such a treatment sequence would allow an SNG plant to meet
the hydrocarbon and sulfur NSPS for gasification proposed by the EPA,
but not more stringent standards such as those proposed by New Mexico.
Best available technology, which would bring the plant into compliance
with the most stringent regulations proposed, would involve removal of
SO from the incinerated tail gases. There are a number of tail gas
treatment systems such as Scot (Shell Development Co.) and Beavon (Union
Oil Co.), but a simpler procedure would be to use an SO scrubber.
The pollution control system shown in Figure VI-3 achieves
both incineration and SO- scrubbing of the sulfur plant tail gas by
routing it to the utility plant. Incineration of the tail gas would
require energy equivalent to several percent of the product SNG, so that
incineration should not be practiced without heat recovery. Incinera-
tion can be achieved most easily by using the oil-fired process heaters
or steam boilers as incinerators. The incinerated tail gas is then
VI-33
-------
routed with the flue gas through the utility plant SC>2 scrubber. The
sulfur plant tail gas is small relative to the utility flue gases, so
the impact on the design of the flue gas treatment system is small. The
treatment sequence is expected to remove more than 98% of the sulfur in
the acid gases (90% in the sulfur plant and 85% of the remainder in the
SO scrubber).
2. Process Condensate Treatment System
The process condensates come from three principal sources:
quenching and scrubbing of the raw synthesis gas to remove oils and ash
carryover; condensation after shifting; and cooling of the synthesis gas
by the Rectisol process. After removal of the ash, the condensates
contain hydrogen sulfide and ammonia, thiocyanates, phenols, aromatic
and fatty acids, acid tars, oil, light hydrocarbons, heavy metals, and
small amounts of suspended solids. The oil fractions can be separated
and removed by skimming, which leaves behind parts per million quan-
tities (solubility limit) of the various components, as well as some
colloidal oil and tar. Many of those compounds are polycyclic aromatic
hydrocarbons, and are considered carcinogenic.
In the Lurgi gasifiers that have been operated commercially,
the condensates contain large amounts of phenols (approximately 5
kg/tonne of coal gasified), which are removed by extraction. The con-
densates from the Synthane demonstration process contain phenols and
water-soluble acid tars that can also be removed by extraction. The
Hygas reactor, however, produces less than half of the phenols that the
Lurgi reactor does. As a result, some full-scale designs for Hygas omit
the extraction step. The phenol concentration in the pooled condensate
stream is only several hundred ppm, which makes it a suitable feed for a
biological treatment system.
Besides phenol, the other principal components of some of the
condensates are H 5 and ammonia, which can be readily removed by steam
VI-34
-------
stripping. Many plant designs call for separating the ammonia from the
sulfide during stripping; the ammonia is sold as a by-product and the
sulfide routed to the sulfur plant for recovery. The stripping opera-
tion also removes many of the volatile hydrocarbons from the conden-
sates; some are recovered as a recycle oil, and many are burned when the
sulfur plant tail gas is incinerated.
The condensate treatment system in Figure VI-3 includes a
solids/water/oil-tar separation step, steam stripping to recover ammonia
and H S and biological treatment. The treated water can be used for
several kinds of recycle and reuse applications. The water is not clean
enough to be used as process water, and could not be economically used
for boiler make-up. It would, however, be suitable for cooling tower
make-up or dust control on mine roads, coal piles, etc., and might even
be useful for revegetation, although continued use as irrigation water
is not recommended because of the probable presence of sulfate and
chloride salts.
Because of the carcinogenic potential of both the heavy metals
and polycyclic aromatics, as well as the potential for the release of
volatile organics through this treatment sequence, Figure VI-4 shows the
treatment sequence in more detail, where it is evident that:
o The sour water stripper removes most of the volatile organics.
o Flocculation and clarification prior to biological treatment
removes many of the heavy metals, much of the colloidal tar
and oil, and polycyclic aromatics, which are highly insoluble.
o Biological treatment removes phenol and most of the bio-
logically degradable organic carbon. Organics with a high
molecular weight, including polycyclic aromatics, are ad-
sorbed onto the biological solids and removed from solution.
Heavy metal levels are reduced to their solubility limit at
pH 7 by adsorption of the metal hydroxide colloids on to the
biological solids.
As a result, the final treated water is well oxidized and
relatively free of heavy metals and organics with high molecular
VI-35
-------
NH3, H2S, VOLATILE ORGANICS
SOUR
CONDENSATES
I
OJ
STEAM
STRIPPER
TO REUSE AND
COOLING TOWER
TREATMENT CHEMICALS
(LIME, POLYELECTROLYTE)
CONDENSATES
-^1 CLARIFIER
ACTIVATED
SLUDGE
HEAVY METALS
SUSPENDED SOLIDS
(ASH, POLYNUCLEAR
AROMATICS, OIL, TAR)
WASTE ACTIVATED SLUDGE
SUSPENDED SOLIDS (POLYNUCLEAR
AROMATICS, OIL, TAR)
HEAVY METALS
FIGURE VI-4. WASTE WATER TREATMENT PLANT FOR COAL GASIFICATION
-------
weights. Release of those compounds to the environment during reuse or
drift and evaporation losses from the cooling tower is small.
3. Solids Processing System
The solids processing system is similar to the design used for
the coal-fired power plant. Dewatered calcium sulfate-calcium sulfite
from the FGS system is mixed with dewatered ash from the gasifier, de-
watered ash from the condensate treatment system, dewatered sludge from
the biological treatment system, and dry fly ash to produce a solids
stream containing approximately 40% water. The solids stream can be
placed in the mine provided it has no contact with surface or ground
waters. If the solids did contact water, they would be a source of
heavy metal and organic carbon pollution.
4. Pollutant Emissions
The major pollutant emissions from the Hygas coal gasification
plant, assuming the controls discussed in the previous section, are
shown in Table VI-10. The emission of regulated pollutants (SO ,
NO , particulates) from the utility plant are all in compliance with
Federal NSPS for coal- and oil-fired boilers. Federal standards for
coal gasification do not exist, although standards have been proposed by
EPA for sulfur and hydrocarbon emissions from the gasification train
(excluding fuel combustion in the utility plant) for first generation
13
Lurgi technology- The emissions shown in Table IV-10 more than meet
those standards by using sulfur plant tail gas incineration and scrub-
bing. The emission of methane/ethane hydrocarbons with the CO- vent
stream is not affected by the proposed regulation because it applies
only to non-methane, non-ethane hydrocarbons. No separate regulations
have been proposed for COS, and the sum of COS and S0_ from the tail
gas does not exceed the proposed sulfur emission standard.
VI-37
-------
No emission of water pollutants is shown in Table VI-10 be-
cause all wastewater is treated and recycled for use within the plant,
resulting in zero discharge of wastewater from the plant site.
Emissions of other pollutants not listed in Table VI-10 are
discussed below.
a. Other Combustion-Related Emissions
The emissions of fine particulates, trace elements, and PAH
from the combustion of coal in the utility plant are similar to those
calculated for the 800 MW coal-fired power plant, because combustion
conditions are similar and we have assumed identical control devices.
The resulting emissions of those pollutants are shown in Table VI-11.
The other source of combustion emissions is the use of by-
product oil as boiler fuel. That oil resembles a light distillate fuel
oil, and has a very high content of C/,+ aromatics. The amount of ash
D
and trace elements contained in the oil are almost impossible to quan-
tify, but are likely to be very small compared to the feed coal. We
assume the oil has a sulfur content of 0.1%. Particulates emitted from
oil combustion are mainly in the form of fine soot resulting from incom-
plete combustion. Such particulates are on the order of a few tenths to
several microns in diameter. Thus, all particulate emissions from oil
combustion are in the fine particulate category (less than 5 microns in
diameter).
The emissions of PAH from various types of oil burners have
been measured. We assume that the results from a steam-atomized
burner (the most commonly used type) apply to the combustion of the
byproduct oil. The higher-than-normal content of aromatics in the oil
may result in a higher content of PAH in the flue gases, but we cannot
quantify the effect.
VI-38
-------
TABLE VI-10
Source
Utility plant
Coal combustion
Fuel oil combustion
Acid gas removal/
sulfur recovery
Tail gas
C02 vent
Coal pile
MAJOR EMISSIONS FROM HYGAS COAL GASIFICATION PLANT
(7.8 MILLION nm3 OF SNG PER DAY)
Emissions Without
Control Devices Control Method
Type
Particulates
so2
NO
x
Hydrocarbons
CO
Particulates
so2
NO
Hydrocarbons
CO
Sulfur
Hydrocarbons
COS
Hydrocarbons
Coal dust
Rate (kg/hr)
3,150
820
45
43
Device
Efficiency (it)
3,550
__ b
14C
6,800d
120e
Precipitator, FGD 99.5
FGD unit 85
Burner design
Burner design
Burner design
FGD unit 50
FGD unit 85
Burner design
Burner design
Burner design
Claus & Stretford
Units 90
FGD 85
Incineration
None
None
None
Emissions Remaining
With Best Control
Rate (kg/hr)
16
120
400
10
32
23
6.4
130
25a
17
110(S02)
14
6,800
120
Coal handling
and crushing
Solid waste disposal
Coal dust
Fly ash
Boiler ash
Gasifier ash
FGD solids
Biological solid
Interstitial wtr
Total
2701
Baghouse
Burial
in
Controlled
Landfill or
Mine
99
Assumes fuel oil combustion/tail gas incineration results in hydrocarbon emissions equal to 50 ppm of
combined fuel oil and tail gas combustion exhaust stream.
Included in hydrocarbon emissions from fuel oil combustion/tail gas incineration.
CAssuming 35 ppm in CO, vent gas.
j ^
Primarily methane and ethane.
eFrom Reference 2 emissions for gravel and aggregate.
From Reference 13.
VI-39
-------
The emissions of fine participates and PAH from oil combustion
are shown in Table VI-11, along with those from coal combustion. The
total combustion-related emission of those pollutants from a coal
gasification facility is also shown.
Table VI-11
COMBUSTION-RELATED EMISSIONS OF FINE PARTICIPATES, TRACE ELEMENTS
AND POLYCYCLIC AROMATIC HYDROCARBONS FROM
THE HYGAS COAL GASIFICATION PLANT
Emissions, g/hr
Pollutant Coal Oil
Fine particulates 8,300 22,700
Trace elements
Sb 0.98 - 0.98
As 0.06 - 0.06
Be 0.36 - 0.36
Cd 2.7 - 2.7
Pd 21.0 - 21.0
Hg 6.3 - 6.3
Se 3.2 - 3.2
Zn 32.0 - 32.0
Polycyclic aromatic hydrocarbons
Fluoranthrene 0.10 0.25 0.35
pyrene 0.22 0.28 0.50
Benzo(a)pyrene 0.02 0.04 0.06
Benzo(e)pyrene 0.03 - 0.03
Benzo(ghi)pyrene 0.01 - 0.01
Benzo(a)anthracene - 1.7 1.7
Phenanthrene - 0.03 0.03
VI-40
-------
b. Other Potential Emissions
The integrated pollution control system illustrated in Figure
VI-3 is designed to concentrate all liquid and solid waste effluent
streams into a single solid waste stream which minimizes the handling
and disposal problems associated with multiple waste streams, many of
which contain toxic materials. Of particular concern are the wastewater
streams that contain such contaminants as cyanides, toxic organics, and
PAH. Although most carbon compounds of concern are likely to be removed
in the biological treatment ponds, PAH are not as susceptible to bio-
logical degradation as other organics, and might pose a health hazard in
the effluent discharge. Toxic trace elements are also of concern in
effluent streams. Many of these are likely to be mobilized into the gas
phase at the high temperatures found in the gasifier, and measurements
of the trace element content of solid residues from Hygas coal gasifi-
cation indicate that this is the case. The fate of those mobilized
elements is not known, although most of them, except for the most
volatile, are likely to be removed in the various condensate streams
downstream of the gasifier. In that case, they would ultimately end up
in a single, concentrated solid waste stream. Some of the more volatile
elements, such as mercury, might end up in the synthesis gas stream, in
which case they could be condensed into the Rectisol cold methanol
stream and be reemitted with the CO vent gas or sulfur plant tail
gas, or they could find their way into the product SNG stream. Such
effects are nearly impossible to quantify until actual measurements are
made on an operating plant. Fugitive emissions of toxic gases may also
be a significant problem, but again, they are nearly impossible to
quantify.
Overall, the most significant environmental control problem
will be the disposal of the concentrated solid waste stream which will
contain nearly all of the toxic and potentially carcinogenic chemicals.
Because of the potential for leaching of those materials from the solid
waste, its disposal must be carefully controlled. Burial in mined-out
areas of the surface mines has been proposed as a solution, but only if
VI-41
-------
there are no nearby aquifiers that can be contaminated. Because of the
occurrence of near-surface aquifers in many coal regions of the West,
more stringent disposal requirements are likely to be imposed under the
Resource Conservation and Recovery Act and other regulations. These
could include the use of controlled disposal ponds lined with clay or
other impermeable materials.
E. Coal Liquefaction Plant
A coal conversion plant that produces liquid hydrocarbons is sim-
ilar in many respects to an SNG plant. Coal is contacted with hydrogen
under conditions of high temperature and pressure to convert it to
liquid and gaseous hydrocarbons. The coal char remaining from lique-
faction is used to produce hydrogen by gasifying it with steam and oxy-
gen in a separate reactor. Therefore, the major difference between a
gasification and liquefaction plant is that during gasification the
reactions of steam, oxygen, and coal to produce hydrogen and the re-
action of hydrogen with coal to produce liquid and gaseous hydrocabons
take place in one reactor, whereas in a liquefaction plant they take
place in separate reactors and under different conditions.
Because the overall reactions are similar, the wastes produced are
similar to those produced in an SNG plant. Table VI-12 summarizes the
major pollutants produced by the various processing operations. The
emission sources listed are similar to those for SNG except that H-Coal
requires a coal drying operation as part of the coal preparation step.
1. Pollution Control System
The important pollution control systems for this plant complex
are the sulfur plant, the flue gas treatment system, the wastewater
treatment system, and the solids handling system.
VI-42
-------
Table VI-12
MAJOR SOURCES OF POLLUTION FOR A COAL LIQUEFACTION PLANT
Medium
Air
Source
Boiler and heater stacks
Emissions
Solid
Water
Coal handling and drying
Boilers and heaters
Water treatment
Hydrogen plant
Sulfur recovery plant
Coal pile
Waste water treatment system
Cooling tower blowdown
Water treatment blowdown
Boiler blowdown
Process condensates
Boiler and hydrogen
plant ash handling
FGD system
Coal storage
SO
x
CO
Hydrocarbons
Particulates (fly ash)
Trace elements
Particulates
S02
NOX
Hydrocarbons
CO
Bottom ash
Fly ash
FGD solids
Suspended solids
Ash and char
Sulfur
Coal fines
Heavy metal hydroxides
Biological solids
Heavy metals
High molecular weight
organic compounds
Dissolved solids
Suspended solids
Heavy metals
H2S
HCN
Ammonia
Thiocyanate
Organic carbon compounds
Suspended solids
Suspended solids
Dissolved solids
Heavy metals
Dissolved solids
Suspended solids
Dissolved solids (heavy
metals)
Suspended solids
Organic carbon compounds
VI-43
-------
The sulfur plant is much simpler than required for SNG produc-
tion because the major gases requiring treatment to remove sulfur are
produced in the H-Coal reactor where only coal, hydrogen, and hydro-
carbons are present. The gas stream produced is not diluted with C02,
and a simple amine scrubber can be used to produce a very concentrated
H S stream. In addition, the conceptual design of the H-Coal plant
uses a different strategy for cleanup of the synthesis gas produced by
the gasifier than was used in the SNG plant design. In H-Coal, H2S
(and some CO ) is removed by an amine scrubber, and the gas is then
shifted to convert CO to CO and steam to hydrogen. Most of the C0_
is removed after shifting. In an SNG plant, on the other hand, the
synthesis gas is shifted first, before any H S is removed; then both
H S and CO are removed in a single step using Rectisol or Selexol.
As a result of the H-Coal strategy, the H S-rich stream (a combination
of the H S streams from the two amine scrubbers) is over 30% H S,
which is a suitable feed for a Glaus plant, and there is no H S-lean
stream requiring treatment in a Stretford plant. Tail gas treatment
includes incineration and lime/limestone scrubbing to control
hydrocarbon, H S, and COS emissions. Incineration is accomplished
with heat recovery using fuel gas produced by the liquefaction reaction,
and scrubbing is accomplished using the utility plant scrubbers. The
flue gas treatment system is the same as was used for the SNG and power
plant cases.
The wastewater treatment plant is similar to the one proposed
for the SNG plant. Process condensates are skimmed to remove liquid
hydrocarbons (tars and oils), settled to remove suspended ash and soot,
and steam stripped to remove H2S and ammonia. The ammonia is recov-
ered as a product and the H^ is sent to the sulfur plant. The water
still contains thiocyanates, phenols, aromatic and fatty acids, heavy
metals, and polycyclic aromatic compounds. The organic material is
probably both qualitatively and quantitatively different from the
organic material produced by a SNG plant because of processing differ-
ences, but very little data exist to confirm this. We have assumed that
no soluble organic materials (phenol in particular) are present in
amounts that can be economically recovered.
VI-44
-------
The wastewater treatment plant is designed to flocculate and
remove suspended solids and heavy metals using lime and organic polymers,
and to biologically oxidize the organic material. After the treatment
sequence of stripping, flocculation, and biological oxidation, the pro-
cess condensates, which are free of heavy metals, polycyclic aromatics,
and volatile organic compounds, are suitable for cooling tower make-up
or dust control in the mine or in coal preparation.
The solids handling system is the same as for SNG. The solids
stream produced for disposal is approximately 40% water, making it easy
to handle. Burial must be in a place suitable for hazardous materials,
because the soluble materials in the solid stream could pollute both
surface and ground waters.
Figure VI-5 shows the integrated water, air, and solid pol-
lution control system. The water flows are approximate and show only
the major water evaporation and water recycle loop. Evaporation losses
in the recycle pond and in the ash quenching operations have been ne-
glected. The cooling tower evaporates water at a rate of 132 liters/sec
(2,100 gal/min), and the scrubber evaporates 7.6 liters/sec (121 gal/min),
The only water discharge is 9.9 liters/sec (157 gal/min) carried away in
the combined solids stream (FGD solids, fly ash, boiler bottom ash,
hydrogen plant slag, and waste activated sludge from biological treat-
ment). If this stream is buried in the mine under conditions where it
contacts neither surface nor underground waters, the plant has zero
discharge of water. (The water balance is not exact, but we constructed
it using reasonable assumptions to demonstrate that zero discharge is
possible for this type of plant.)
The neglected streams include approximately 25 liters/sec
(400 gal/min) of boiler blowdown, which will be balanced by evaporation
during quenching operations and evaporation from the recycle pond.
Additional water inputs to the plant include approximately 76 liters/sec
(1,200 gal/min) for dust control on roads and in the mine 95 liters/sec
(1,500 gal/min ) boiler feed water, and 19 liters/sec (300 gal/min) for
VI-45
-------
<
M
RAW WATER
250 LPS -
PROCESS
BOILER FEED
MINE USE
6 CYCLES
I EVAPORATION
AND DRIFT
132 LPS
WATER I 59 LPS I COOLING
TREATMENTH TOWER
COOLING TOWER
MAKEUP
BLOW DOWN
18 LPS
COAL
C1-C4 HYDROCARBONS
161 LPS
DISTILLATE
PRODUCT
GAS
NH3
T ] CLEAN UP
1
CONDENSATE
'Y OIL
*
5EN GA
r " CLEAIV
*
UP|
OIL- WATER
SEPARATION
I
H
SULFUR
PLANT
T
^ AMMONIA 1
RECOVERY
2S
CONDENSATE
TAIL GAS
SCRUBBER MAKE UP 35 LPS
LPS
FGD SLUDGE 28 LPS
FLY ASH
ASH SLURRY
ASH AND CHAR SLURRY
93 LPS
1 BIOLOGICAL
TREATMENT!
RECOVERED
WATER
144LPS
BIOLOGICAL SOLIDS
SOLIDS
CONCENTRATION
I COAL PILE
RUNOFF
NEUTRALIZATION
LPS: LITERS PER SECOND
9.9 LPS
TO SOLIDS
DISPOSAL
CLARIFIER
TO RECYCLE POND
FIGURE VI-5. INTEGRATED POLLUTION CONTROL SYSTEM FOR AN H-COAL LIQUEFACTION PLANT THAT
PRODUCES 7950 tn3 PER DAY OF FUEL
-------
showers, washrooms, and miscellaneous process uses, resulting in a total
water consumption of 250-320 liters/sec (4,000-5,000 gal/min).
2. Pollutant Emissions
The major pollutant emissions for an H-Coal liquefaction fa-
cility are summarized in Table VI-13. As in the case of the coal gasif-
ication plant, the control devices for coal and by-product gas combus-
tion in the utility plant will meet or exceed current NSPS. There are
/
currently no standards or proposed standards for coal liquefaction.
However, the level of controls specified for sulfur recovery and subse-
quent tail gas treatment should more than meet any likely standards.
No water pollutant emissions are shown in Table VI-13 because
the treatment system has been designed to achieve zero discharge.
a. Other Combustion-Related Emissions
The combustion of coal and by-product gas in the utility plant
results in the emission of fine particulates, trace elements, and PAH.
The emissions from coal combustion are analogous to those from the coal-
fired power plant. For gas combustion, no significant quantities of
trace elements should be emitted. However, fine particulates resulting
14
from incomplete combustion and some PAH are emitted. . The emissions
of those pollutants from both coal and gas combustion, and the totals
from the two sources, are shown in Table VI-14.
b. Other Potential Pollutants
As in the case of coal gasification, most of the effluent
streams containing toxic waste products are combined and concentrated
into a single solid waste stream that must be carefully disposed of.
VI-47
-------
TABLE VI-13
MAJOR EMISSIONS FROM H-COAL LIQUEFACTION
(7,950 m3 OF DISTILLATE FUEL OIL PER DAY)
Emissions Without
Control Devices
Control Method
Emissions Remaining
With Best Control
Source
Utility plant
Coal combustion
Fuel gas combustion
Coal dryer
Acid gas removal/
sulfur recovery
Coal pile
Coal handling & crush
Solid waste disposal
Type Rate (kg/hr)
Particulates
S°2
NO
X
Hydrocarbons
CO
Particulates
SO,
NO
X
Hydrocarbons
CO
Coal dust
Sulfur
Hydrocarbons
Coal dust
Coal dust
Fly ash
Boiler ash
FGD solids
Hyd. plant slag
Biological Solid
Interstitial wtr
Total
7«,690
2,000
990
24
80
7.3
—
210
2.2a
12
6,320
4,390
b
180C
410d
7,700
1,900
5,200
58,300
1,800
35,800
110,700
Device Efficiency (Z)
Precipitator
FGD
Burner design
Burner design
Burner design
FGD
—
Burner design
Burner design
Burner design
Multiple cyclones
& venturi scrubber
Claus Unit
FGD
Incineration
None
Baghouse
Burial
in
Controlled
Landfill
or Mine
99
85
—
—
—
50
—
—
—
—
99
95
85
—
—
99
Rate (kg/hr)
39
300
990
24
80
3.6
—
210
2.2
12
63
66(S02)
__ b
180
4.1
7,700
1,900
5,200
58,300
1,800
35,800
110,700
COTnp°nent in flue 8as' Hydrocarbon emission is the standard factor for
blncluded in emissions from fuel gas combustion/ tail gas incineration.
cFrom Reference 2 emissions for gravel and aggregate.
dFrom Reference 13.
VI-48
-------
Although no quantitative measurements have taken place, the amount of
higher molecular weight organics produced in coal liquefaction are
likely to be greater than in coal gasification, because the process con-
ditions for coal liquefaction are designed precisely for producing a
wide range of liquid hydrocarbons, while coal gasification process con-
ditions are designed to maximize the production of methane, carbon mon-
oxide, and hydrogen. Many types of PAH are likely to be found in the
coal liquefaction product, which contains a large fraction of aromat-
ics. There is some concern over the release of those compounds during
combustion in various end use applications. The effects have not been
quantified, however.
Table VI-14
COMBUSTION-RELATED EMISSIONS OF FINE PARTICULATES, TRACE ELEMENTS
AND POLYCYCLIC AROMATIC HYDROCARBONS FROM AN H-COAL LIQUEFACTION PLANT
Emissions (g/hr)
Pollutant Coal Gas Total
Fine particulates 20,100 3,600 23,700
Trace elements
Sb 2.5 - 2.5
As 0.14 - 0.14
Be 0.94 - 0.94
Cd 6.9 - 6.9
Pb 16.0 - 16.0
Hg 53.0 - 53.0
Se 8.1 - 8.1
Zn 81.0 - 81.0
Polycyclic aromatic hydrocarbons
Fluoranthrene 0.25 0.26 0.51
Pyrene 0.57 0.35 0.92
Benzo(a)pyrene 0.06 0.02 0.08
Benzo(e)pyrene 0.07 0.04 0.11
Benzo(ghi)pyrene 0.02 - 0.02
Coronene - 0.03 0.03
Phenanthrene - 0.18 0.18
VI-49
-------
The effluent stream with the largest concentration of organics (and
probably trace elements) is the sour process condensate from the H-Coal
reactor. It is treated in an oil separation unit, and steam stripped
for removal of ammonia and H S before being sent to the biological
treatment ponds. Ultimately, the residue from that stream is combined
with the solid waste stream for disposal and the treated water will be
recycled to the cooling towers.
Other air emission sources around the plant include fugitive
emissions from high pressure vessels and piping and emission of hydro-
carbons (mostly methane and ethane) with the CO vent streams from
acid gas removal in the hydrogen plant. Such emissions have not been
quantified, however.
F. Pipelines
The environmental factors associated with pipeline construction and
operation are physiography, hydrology, vegetation, wildlife, and air
quality. Within the context of each of these divisions, the various
impacts of construction, operation, and maintenance are discussed below.
1. Physiography
The types of soil found within the various regions crossed by the
pipeline routes are important because the pipelines are constructed
below ground. Construction actions that create temporary or permanent
soil alteration include access road construction, vegetation clearing,
trenching, blasting, materials storage/stockpiling, equipment movement,
and backfilling. The operation and maintenance activities that affect
soil conditions are access road maintenance, regulation control, and
facility repair. The environmental impacts of most concern are com-
paction and subsidence, slope stability, and surface erosion.
VI-50
-------
During the right-of-way construction processes, the vegetation is
removed and the soil is laid bare. Potential revegetation of this
stripped area is highly variable. Prior to regrowth, the right-of-way
strip approximates a continuously fallow condition and is much more
sensitive to erosion than the natural state. The area subject to damage
is determined by the width of the construction right-of-way, which
typically is 30 m (100 ft.).
Heavy equipment movement has an adverse effect on the soils by
creating gullies and subsequent erosion. Where regional slope erosion
has not been a problem in the past and where revegetation or land use
has precluded erosion, no serious effects are apt to occur.
The stream and river bank erosion at river crossings depends on the
nature of the river and the composition of the bed and bank materials.
However, because problems of this nature generally have engineering
solutions, mitigation is generally possible.
Carefully planned pipeline construction usually does not cause or
add to existing slope stability problems. However, two general cate-
gories of possible slope stability hazards must be considered: (1)
existing unstable slopes whose future stability or instability depends
on specific soil structure and slope details; and (2) some steep slopes
not involved in past sliding action may, nonetheless, be unstable and
therefore potential slide areas.
However, soil slide masses exist and should be considered poten-
tially troublesome ground during construction. Many old soil slides
continue to move downhill intermittently or, in some cases, contin-
uously, in a slow, glacier-like fashion. If the rock and soil composing
the slide mass are relatively loose and fragmented, the potential for
slide tends to increase.
Other problematical soil conditions are those soils with a high
water table and those with low fertility. Both of these conditions pose
VI-51
-------
environmental problems if disturbed. The soils with high water tables
have a high potential for adverse change in the physical and biological
environment. Disturbance of very low fertility soils has adverse impli-
cations for the speed of revegetation.
2. Hydrology
The potential for creating environmentally adverse effects are high
when pipelines are constructed across streams, rivers, and lakes.
Pipeline construction, operation, and maintenance can have a detrimental
effect on existing water resources, and thus on vegetation cover, wild-
life habitat, and the recreation. Construction activities that can
affect hydrologic resources are access roads, access canals, vegetation
clearing, trenching and dewatering, blasting, equipment movement, dredg-
ing, structures, and waste disposal. Operation and maintenance activi-
ties that can impact hydrologic resources are access road maintenance,
access canal maintenance, vegetation control, liquid release, and faci-
lity repair.
Areas of major environmental concern include lakes and rivers or
stream crossings, surface water conditions, and groundwater. Crossings
above the high water level are not aesthetic. In addition, the con-
struction of a pipeline on fill and/or pier supports in a lake or river
adversely affects the water quality by increasing water turbidity, thus
reducing the value of the lake or river as a recreational resource.
Another impact, not as severe, occurs when pipelines and access roads
are constructed near lake or river shores.
Existing surface water conditions affect the pipe and its main-
tenance requirements. Excessive turbidity downstream reduces the
penetration of light into the water and reduces the photosynthetic
activity of submerged vegetation. Excessive sediment loads in streams
having gravel or rubble-type bottoms are damaging to trout and salmon
because the sediment fills the interstices of the gravel or stones on
the streambed, thus eliminating spawning grounds.
VI-52
-------
Dewatering of trenches and other excavations is difficult where the
water table is near the ground surface and aquifer permeability and
porosity is high. Dewatering operations can cause significant effects
on aquifer compaction (which may lead to ground subsidence), inter-
ference with water supplies, and changes in the water table affecting
vegetation and wildlife habitat near the right-of-way.
3. Vegetation
The type of vegetation a pipeline right-of-way crosses is extremely
important in its influence on wildlife habitat, economic resources,
visual perception and land use. Removal or destruction of vegetation
related to the constuction, operation, and maintenance phases is one of
the most obvious environmental impacts, particularly because of the
complex relationships between vegetative cover and the other factors.
Construction activities that lead to temporary or permanent
destruction or alteration of vegetation are the creation of access
roads, vegetation clearing, trenching, materials storage/stockpiling,
equipment movement, backfilling, structures, and waste disposal. Oper-
ation and maintenance activities that affect the vegetative cover are
access road maintenance, vegetation control, facility repair, and
right-of-way abandonment. Specific impacts resulting from pipeline-
related activities are dependent on the type of vegetation crossed by
the pipeline (natural resource areas, wetlands, forestlands, and
agricultural land).
Pipeline route configurations should avoid areas of designated
unique natural resources valuable for observation and/or scientific
research because of their relatively pristine quality. The mere
physical presence of a cleared pipeline right-of-way has substantial
detrimental impact on this quality.
Construction of pipelines in forested regions requires the complete
removal of vegetation from the right-of-way. This action effects an
VI-53
-------
immediate and long-term change in the beauty of the forest; it also
exposes bare soil to wind and water erosion with adverse impact on soil
resources and water quality.
Rights-of-way through grasslands or brushlands with scattered trees
(woodland) normally has minor impact when planned in accordance with
existing tree patterns. However, in areas of denser tree cover,
extensive clearing adversely affects the woodland habitat.
When pipelines are routed through cultivated land, pasture, and
range land, the main impact is the permanent loss of crops from land
required for compressor station structures and permanent access roads.
Periodic maintenance or emergency repair by way of temporary paths
across fields also results in crop losses. However, the impact on
agricultural resources is usually low because the original soil
fertility can be established after burial of the pipe and normal crop
production can resume.
4. Wildlife
Adverse impacts to wildlife and wildlife habitat vary throughout
any region because of varying habitat type and range and diverse methods
of construction, operation, and maintenance. In general, disruption of
the terrestrial wildlife by a project reaches its highest point at
certain vulnerable locations along the route during the construction
phase. In vast areas of single habitat type surrounding a pipeline
right-of-way, the relative effect is low. In some locations, some
reclamation may be required.
The presence of compressor stations and mainline valves along a
right-of-way permanently eliminates the prior vegetation and wildlife
from the site. Much of the land surface of these properties is covered
with concrete, gravel, or asphalt and installations are typically en-
closed with chain link fences. Compared to the volume of adjacent
VI-54
-------
habitat, this effect is comparatively minor, although the impacts are
absolute and irreversible.
The impacts on terrestrial wildlife of operation of the compressor
stations are not sufficiently known. However, compressor stations are
noisy and in the interest of avoiding wildlife disturbance should be
acoustically treated.
The temporary gross physical disturbance and siltation of aquatic
habitat at river crossings is probably the most serious problem to pre-
vent during construction. Also, during the testing phase of pipeline
construction, large amounts of water may be discharged in a relatively
short time into water courses, temporary holding ponds or onto nearby
ground that drains into the water courses along a right-of-way. The
timing and magnitude of these releases can affect aquatic populations.
Streamside slash and other debris that may drift into the water
courses crossed by a pipeline right-of-way requires adequate disposal.
Organic debris can block streams and prevent or curtail fish movements
and migrations and cause disruption of spawning.
After construction, a pipeline, properly buried under several feet
of sand, gravel, and varying depths of water, exerts no discernible ef-
fect on either the character of the refilled bottom or on water quality,
quantity or rate of flow. Potential impacts on the water resources,
similar to those from construction, would stem from repair of the
pipeline.
5. Air Quality
The major impact on air quality by pipeline operations is the
emissions from the compressor or pumping stations distributed over the
1,300 km (800 mi) length of the pipeline. The fuel consumption by these
stations was discussed in Chapter V.
VI-55
-------
The emissions from the turbine-powered compressor stations on the
natural gas pipeline are shown in Table VI-15 . For the pipeline in the
system we assumed, a total of 11 compressor stations are spaced at
intervals of approximately 110 km (67 mi).
The air pollutant emissions from the diesel-powered pumping
stations on the liquid fuels pipeline are shown in Table VI-16. Ten
such stations are spaced at intervals of approximately 130 km (80 mi)
along the 1,300 km pipeline we assumed.
Because the emissions are so dispersed, they do not create a
significant air pollution problem. However, when compressor or pumping
stations are located near urbanized areas, the emissions add to the
pollutant loading generated by automobiles, industry, and other sources.
G. Fuel Distribution
1. Tank Trucks
Dispersed 26-MW fuel-cell power plants will sometimes be con-
structed in residential areas. Naphtha-fueled power plants will
probably be supplied by tank trucks. The trucks designed for such short
hauls are normally straight, rigid-bodied trucks able to service a
decentralized market.
Most tank trucks today are diesel-powered because less maintenance
is required and because diesel fuel is less expensive than gasoline.
Modern tank trucks have become progressively lighter and stronger
because of aluminum alloys, stainless steel, and reinforced fiberglass
which is now used in the construction of the tank compartments instead
of carbon steel.
VI-56
-------
Table VI-15
AIR POLLUTANT EMISSIONS FROM A COMPRESSOR
STATION ON AN 81 CM (32 IN.) NATURAL GAS PIPELINE
Pollutant Emissions, kg/hr (Ib/hr)
NO 33.0 (72.0)
A.
CO 13.0 (29.0)
Hydrocarbons 2.5 (5.5)
S00 0.065 (0.14)
*Non-methane hydrocarbons represent 5 - 10% of the total,
Source: Reference 2.
Table VI-16
AIR POLLUTANT EMISSIONS FROM A PUMPING STATION
ON A 51 CM (20 IN.) LIQUID FUELS PIPELINE
Pollutant Emissions, kg/hr (Ib/hr)
N0x 44.0 (98.0)
CO 9.6 (21.0)
Hydrocarbons 3.5 (7.8)
Particulates 3.2 (7.0)
SO 2.9 (6.5)
A,
Aldehydes 0.67 (1.5)
Source: Reference 2.
VI-57
-------
Tank trucks are designed mainly for travel along highways and
thoroughfares and not residential streets. When these trucks do utilize
such streets their presence is very noticeable because of their size and
noise levels. Tank trucks used to deliver fuel to fuel-cell facilities
would create impacts such as air pollution, safety hazards, and the pos-
sibility of damage to the surface of the street due to the weight of the
trucks.
Tank trucks are substantially larger than the average automobile.
Because of their size they have aesthetic impact. The significance of
the impact will depend on the time of day in which deliveries are made,
the frequency of deliveries, and the appearance of the truck itself.
Trucks are a safety hazard mainly due to their size. They are more
damaging when they collide with another vehicle or a person than the
standard automobile which normally travels through the community.
The safety hazard these trucks pose in a community is dependent on
the times of delivery and the makeup of the neighborhood — whether
there are children playing in the street, older people who are no longer
very agile, or whether it is mainly a community of young and middle aged
people who are generally working during the day and would for the most
part be unaware of the tank trucks.
The most noticeable and well studied impacts of trucks is their
noise level. The noise impact is the most significant environmental
impact resulting from fuel distribution. To provide an idea of the
contribution of each source to the overall noise created by a diesel
truck, Table VI-17 presents a weighted sound level at 15 m (50 ft.) The
noise levels increase with vehicle speed and also depend upon variables
such as the road surface, axle loading, tread design, and wear condi-
tion. Change in any of the variables can result in variations in noise
level of up to 20 dB at constant vehicle speed. Truck tires are gener-
ally noisier than automobile tires because of their size and other
VI-58
-------
design constraints. Engine-generated noise normally dominates at speeds
below 45 mph. This noise radiates directly from the engine exhaust and
intake openings and from vibrating engine casing. Other major sources
of truck noise are turbulent aerodynamic flow over the body and the
rattling of loose mechanical parts.
Diesel truck noise is considered a nuisance as well as disruptive;
it is termed "intermittent single-event" noise. Generally, this type of
noise interferes with speech and other activities for brief intervals.
The impact this noise will have on different communities will depend on
the residual noise level, and as can be seen in Table VI-18 there are
differences in the noise levels found in even residential communities.
Table VI-17
DIESEL TRUCK NOISE SOURCES
Source Weighted Sound Level at 15m, dB(A)
Engine 78
Exhaust 85
Intake 75
Fan 82
Tires 75 (less than 58 km/hr)
95 (greater than 58 km/hr)
Total 88 (less than 58 km/hr)
96 (greater than 58 km/hr)
Source: Reference 16.
VI-59
-------
Table VI-18
URBAN AND SUBURBAN DETACHED HOUSING RESIDENTIAL AREAS
AND APPROXIMATE DAYTIME RESIDUAL NOISE LEVEL (L9Q)
Neighborhood Typical Range
Description dB(A) Average dB(A)
Quiet suburban residential 36 to 40 38
Normal suburban residential 41 to 45 43
Urban residential 46 to 50 48
Noisy urban residential 51 to 55 53
Very noisy urban residentiaL 56 to 60 58
Source: Reference 17.
Finally, the effect of these tank trucks on the air quality of the
community is important. Currently, about 54 percent (by weight) of all
air pollution in the United States is emitted by mobile sources and
between 50 and 75% of the total weight of three pollutants — hydrocar-
bons, carbon monoxide, and nitrogen oxide — come from transportation.
Air pollution affects humans, animals, plants, materials, and visibil-
ity. The impacts from the air pollution emitted by automobiles has had
significant detrimental impacts. Diesel trucks add pollutants to the
air of any community through which they travel, but when compared to the
already existing problem this impact can be considered insignificant as
a pollutant source. Table VI-19 provides a summary of diesel truck
emission.
2. Unit Train (Fuel Oil)
The impacts of the unit train supplying distillate fuel oil to a
combined cycle power plant are generally similar to those of the coal
unit train. However, due to the lower weight of a 100-tank car unit
VI-60
-------
train, its emissions would be correspondingly less. The air pollutant
emissions for the coal unit train were shown in Table VI-3. The loaded
coal train data should be multiplied by 0.52 to obtain the emissions
from a tank car unit train carrying distillate fuel oil; the round trip
averages should be multiplied by 0.63.
Table VI-19
AIR POLLUTANT EMISSIONS FROM
A DIESEL-POWERED TANK TRUCK
Pollutant Emissions
g/km Ib/mi
co* 18.0 (0.063)
Hydrocarbons* 2.9 (0.010)
NOX * 11.0 (0.040)
SOX ** 1.7 (0.006)
Particulates 0.81 (0.003)
Aldehydes 0.2 (0.0006)
*EPA estimates for 1990 model year vehicles.
**Assumes fuel sulfur content of 0.2 %.
H. Combined Cycle Power Plant
Because the combined-cycle power plant burns low-sulfur distillate
fuel, no stack gas emission controls are necessary. The most serious
emisson problem is the large amount of NO produced in the
high-temperature combustor. The various methods of controlling the
formation of NO were discussed in Section IV-D. The H-Coal
distillate fuel contains 0.23% bound organic nitrogen (see Section
IV-C). Combustion tests have shown that approximately one-half of bound
nitrogen in fuel is converted to N0x in the combustor, resulting in an
emission level of about 0.065 kg NO /GJ (0.15 Ib NO /10^ Btu) for
x x
H-coal distillate fuel. That means that fixation of nitrogen from the
VI-61
-------
feed air must yield no more than 0.065 kg NO /GJ (0.15 lb NO /10 Btu)
6 X
to meet a possible NSPS of 0.13 kg NO /GJ (0.3 lb NO /10 Btu). The
X X
use of proper burner design, water injection, and staged combustion has been
shown to control NO to low levels in conventional gas turbines operating
X
at 1,000-1,100°C (1,850-2,000°F). Whether such low levels can be
achieved in a 1,370°C (2,500°F) turbine is not known. However, for the
purpose of analysis, we assume that through water injection and other
control measures the NSPS for NO is met.
X
The emission of SO was calculated on the basis of a sulfur con-
tent in the H-Coal distillate product of 0.11%. The emissions of other
pollutants were calculated using emission factors for oil-fired tur-
bines. The major emissions to the air from the 270-MW combined-cycle
power plant are shown in Table VI-20.
Table VI-20
EMISSIONS FROM A 270-MW COMBINED-CYCLE POWER PLANT
Source Type Emissions (kg/hr)
Stack gases
Cooling tower blowdown,
boiler blowdown,
and fuel preparation
particulates
so2
N0a
x b
Hydrocarbons
cob
Suspended solids'
t f*
Oil and grease
Iron
Copper0
Chlorine0
48.0
91.0
240.0
23.0
23.0
5.6
2.8
0.2
0.2
0.003
aAssumes that NSPS of 0.3 lb NOX/106 Btu fuel is met.
bSource: Reference 18.
GBased on effluent limitation guidelines in Reference 5.
VI-62
-------
No solids handling facilities are required for the combined-cycle
plant because no solid wastes are generated in its operation. The main
source of wastewater is the cooling tower blowdown, which contains small
amounts of dissolved solids and heavy metals. In addition, a small
discharge of water from the fuel treatment section contains dissolved
salts of sodium and potassium in addition to some suspended oils. All
wastewater is assumed to receive appropriate treatment prior to
discharge.
The calculation of toxic trace element and PAH emissions from the
combined-cycle power plant is difficult because of a lack of knowledge
of the trace element concentration in the H-Coal distillate fuel and the
potential of this fuel to form PAH in the high-temperature combustor.
However, we assume that the trace element content of H-Coal liquids will
be at least an order of magnitude lower than in the feed coal. This
assumption results from measurements of the trace element content of SRC
and COED process liquid products relative to the feed coal from which
19
they are derived. Reduction in concentration varied with the parti-
cular trace element and the process. However, an overall reduction fac-
tor of about 10 represents a reasonable average. Using that factor, and
the assumption that all trace elements contained in the fuel oil are
emitted with the stack gases. The emissions of trace elements from the
270 MW combined-cycle power plant are shown in Table VI-21.
Table VI-21
EMISSIONS OF TOXIC TRACE ELEMENTS FROM
A 270-MW COMBINED-CYCLE POWER PLANT
Element Emissions (g/hr)
Sb 3.0
As 13.5
Be 3.2
Cd 9.5
Pb 32.0
Hg 0.45
Se 3.3
Zn 150.0
VI-63
-------
The only reasonable assumption that we can make regarding the emis-
sions of PAH from the combined-cycle plant is that they will be similar
to those from conventional oil-fired boilers. Extensive testing of com-
bustion products must be performed before it is known whether combustion
of coal-derived fuels will result in substantially higher PAH emissions.
Estimation of PAH emission based on analogy with conventional oil com-
bustion is shown in Table VI-22.
I. 26-MW Fuel-Cell Power Plant (SNG)
Emissions from the System 2 power plant were estimated based on
rated-load operation. Under normal conditions, only one process stream
leaves the power plant (Stream 33). The composition of this stream is
shown in Table VI-23.
Production of NO pollutants in the combustion zone of the
X.
reformer furnace was assessed using the equilibrium and kinetic model
procedures described later in Section VI-M. The results, shown in Table
VI-24 indicate negligible NO production.
X
Table VI-22
EMISSION OF PAH FROM A 270-MW COMBINED-CYCLE POWER PLANT
Compound Emission (g/hr)
Pyrene 0.55
Fluoranthene 0.50
Phenanthrene 3.3
Benzo(a)pyrene 0.087
Benzo(a)anthracene 0.050
VI-64
-------
Table VI-23
COMPOSITION OF EFFLUENT PROCESS STREAM
FOR A NOMINAL 26-MW FUEL-CELL POWER PLANT
Emission Rate
Component at Rated Load (kg/hr)
C02 9,140
H20 (vapor) 8,270
02 2,940
N2 58,100
CO Nil*
CH4 Nil*
NO See Table VI-24
X
Particulates Nil
Waste heat to atmosphere 116 GJ/hr
Based on equilibrium combustion calculations.
Table VI-24
PREDICTED REFORMER FURNACE NO,, PRODUCTION
A
Residence Time3 Predicted NOV Levels^
(msec) Mole Fraction Equivalent gNOv/hr
5 6.0 x 10~13 4.66 x 10~5
10 1.2 x 10~12 9.32 x 10~5
100 1.2 x 10"11 9.32 x 10~4
aAdiabatic flame temperature = 1,161°C (2,121°F)
rated load, assuming NOX molecular wt = 30.
VI-65
-------
The power plant has little effect on the environment. The major
effluents are water, CO and thermal energy, which are dissipated
directly into the atmosphere. The system is water-conservative, re-
quiring no make-up water for its operation. Thus, it adds no thermal
pollution to rivers, lakes, or streams.
Power plant operation does involve the generation and flow of
hydrogen-rich mixtures. Leakage could result in explosions; however,
proper component design and fabrication procedures minimize this
potential hazard.
Other pollution aspects of the power plant are reviewed below.
1. Spent Fuel-Cell Stack Disposal
The molten carbonate fuel-cell stacks are assumed to have a
nominal working life of 40,000 hours. Periodic replacement is required
during the operating life of the power plant. Because molten carbonate
cells do not use expensive noble metal catalysts, stack recovery is not
required. Rather, suitable disposal procedures can be expected to be
developed to handle the discarded stacks. Few problems should arise,
even though replacement of all power plant stacks involves discarding
about 87 tonnes (96 tons) of lithium aluminate tile plus mixed
lithium/potassium carbonate electrolyte. Disposal of the remaining
stack components, including sintered nickel electrodes and stainless
steel structural members, should be straightforward.
2. Spent Catalyst Guard Bed Disposal
The power plant contains several catalyst and adsorbent beds
that require periodic replacement. The resulting pollution burden is
small. For example, the spent sulfur guard bed involves removal of
about 0.018 kg (0.04 Ib) of zinc sulfide per MWh of operation. This is
VI-66
-------
equivalent to 1,310 kg/yr (2,880 Ib/yr), assuming 3,000 hours annual
operating time at full load. Spent reforming catalyst (nickel) could be
returned to the supplier for safe disposal.
3. Noise
Noise levels generated by the power plant have not been
defined. The plant contains rotating machinery that will have charac-
teristic noise signatures. Scaled-up phosphoric acid power plants are
expected to generate 55 dBA at 30 m (100 ft).20 Similar levels should
apply to the molten carbonate power plant.
J. 26-MW Fuel-Cell Power Plant (Naphtha)
Emissions from the System 3 power plant, operating at rated load,
were estimated, using the bases discussed in Section VI-I. As shown in
Table VI-25 and VI-26, the power plant has little impact on the environ-
ment. Although the naphtha feed contains sulfur, no sulfur oxide
emissions are shown in Table VI-25. The fuel conditioning section
contains a hydrodesulfurization unit and zinc oxide guard bed to adsorb
H S. Replacement of spent guard beds involves the removal of about
0.014 kg (0.03 Ib) of zinc sulfide per MWh of operation. This is
equivalent to 900 kg/yr (1,980 Ib/yr), assuming 3,000 hours annual
operating time at full load.
The target sulfur level in the naphtha feed to the reformer is 0.2
ppm sulfur. We assume that this sulfur, after conversion to H s in
the reformer, is retained on the nickel reforming catalyst. There is
also increasing evidence that small quantities of sulfur may be retained
within the molten carbonate fuel cell itself.
VI-67
-------
Table VI-25
COMPOSITION OF EFFLUENT PROCESS
STREAM FOR A NOMINAL 26-MW FUEL-CELL POWER PLANT
Component Emission Rate at Rated Load (kg/hr)
CO 13,100
H20 (vapor) 6,040
0 3,250
N. 65,100
*
CO Nil
*
CH, Nil
NO See Table VI-26
x
Particulates Nil
Waste heat to
atmosphere 111 GJ/hr
Based on equilibrium combustion calculations.
Table VI-26
PREDICTED REFORMER FURNACE NOX PRODUCTION
Residence Time3 Predicted N0y Levelsb
(msec) Mole Fraction Equivalent (gNOY/hr)
5 4.8 x 10~12 3.92 x 10"4
10 9.6 x 10~12 7.83 x 10"4
100 9.6 x 10~U 7.83 x 10"3
aAdiabatic flame temperature = 1,225°C (2,236°F)
bAt rated load, assuming NOX molecular wt = 30.
VI-68
-------
Other environmental factors are similar to those of the SNG-fueled
power plant, discussed in the previous section. The naphtha-fueled
power plant is also water-conservative and does not generate water-borne
thermal pollution. Spent fuel-cell stack disposal in this case involves
discarding about 69 tonnes (79 tons) of mixed electrolyte plus ceramic
tile. Environmentally acceptable means must be developed for disposal
of spent catalysts, including:
o Hydrodesulfurization (Ni-Mo)
o Reforming (Ni)
o Shift conversion
Lastly, noise pollution characteristics for the naphtha-fueled power
plant remain to be defined. However, they are expected to be similar to
scaled-up phosphoric acid power plants, as discussed in the previous
section.
K. Electricity Transmission and Distribution
Meeting the increasing demand for electric energy requires not only
additions to the energy generating capacity, but also additions to the
capacity to move the electric power from the location of the generating
plant to the dispersed users of the electricity. Power transmission
lines of increasing length and voltage capacity are becoming more impor-
tant as new generating plants are located farther from the final energy
users than older plants. New and planned transmission lines will inter-
connect electric systems in different areas to improve the reliability
and level of service. Today's transmission lines carry power at higher
voltages, on the average, than those existing a decade ago. The trend
toward higher voltages is expected to continue as research extends the
technical limits of transmission line capacity. However, research is
required to solve some of the increased environmental impacts expected
of the higher capacity lines.
VI-69
-------
1. Transmission Line Characteristics
a. Voltages
At present, transmission lines operate at voltages of between
115 and 800 kV. Lines operating at voltages below 115 kV are considered
subtransmission and distribution lines. Future line voltages of 1,200
and 1,500 kV are the subject of current research and development.
b. Towers
The size and design of transmission lines towers are primarily
dependent on the voltage of the line and number of circuits to be car-
ried. The height of the towers and the minimum line sag determines the
spacing between the towers. Thus, higher towers can be spaced farther
apart and carry the same voltage.
c. Right-of-Vay
All overhead transmission lines require a right-of-way rela-
tively clear of vegetation such as tall trees, which could fall into the
towers or lines, and other vegetation, which could hamper movements of
the people who maintain the line. The width of the right-of-way is gen-
erally proportional to the voltage of the line, but many other factors
such as number of circuits, topography, vegetation, surrounding land
use, type of land purchase or lease, and land values also determine the
width of the right-of-way. For example, a 345-kV transmission line
sited in a flat agricultural area might require a 30 m (100 ft) right-
of-way, while the same size line in a rugged, forested terrain might
require a 76 m (250 ft) right-of-way. Table VI-27 shows average
right-of-way widths for different voltage lines.
VI-70
-------
Table VI-27
AVERAGE REQUIREMENTS FOR RIGHTS-OF-WAY
Voltages
(kV)
345
500
765
1,200
Right-of-Way
Width, m (ft)
46 (150)
61 (200)
76 (250)
91 (300)
2. Transmission Line Impacts
a. Electrical Impacts
Impacts related to the electrical characteristics of trans-
mission lines fall into two categories: impacts related to electric
fields surrounding the lines and those related to corona discharge.
Electric fields are known to induce voltages in objects under the lines,
which can result in unpleasant or possibly hazardous (under rare condi-
tions with children) electric shock. A few experts suggest that 60 Hz
electromagnetic fields are hazardous to people and animals, although
experts do not generally agree on what are the biological effects of
fields produced by transmission lines.
Corona and electric fields at ground level both increase
directly with increase in operating voltage of a transmission line.
Existing transmission systems cause little or no concern over electrical
impacts because operating voltages have been below 500 kV until very
recently. However, high voltage transmission systems are now being
designed at voltages up to 765 kV and research is being done on 1,200
and 1,500 kV-lines. Thus, controlling the impacts of corona and
electric fields is increasingly important in designs of present and
future transmission systems.
VI-71
-------
1) Corona and Related Impacts—Corona, or corona discharge,
refers to a luminous discharge (due to ionization of the air surrounding
a conductor) caused by the voltage gradient exceeding the breakdown vol-
tage of air. The discharge yields heat, light, audible noise, electri-
cal static, and vibration. Irregularities on the conductor surfaces
such as water drops are many times more conducive to corona discharges
than are clean, dry conductors. Thus, corona discharges predominantly
occur during wet weather such as rain, heavy snow, or fog.
Audible noise (AN) is produced by corona discharges from
a transmission line. Because of this association, AN during wet weather
is much louder than during dry weather. In fact, dry weather levels,
which may average 35 dB for a 765-kV line, are generally masked by
background noise and therefore are not a cause for concern. During rain
or heavy snow, a 765-kV line will produce a noise level of about 56
dB(A) measured 38 m (125 ft) from centerline (the approximate edge of
the right-of-way). Because AN is roughly proportional to the operating
voltage of the transmission line, a line operating at less than 765 kV
will generally produce less than 56 dB(A) under identical conditions of
rain or snow. Although 56 dB(A) is below the safety threshold for
hearing damage (70 dB), it is high enough to interfere with sleep, mask
speech, and cause general annoyance.
Sleep interference becomes a potential problem when
bedroom noise levels reach 33-35 dB(A). Residences located at or very
near the edge of the right-of-way might experience, during wet weather
with windows partly open, indoor noise levels between 34 and 43 dB(A)
and 23 dfi(A) with windows closed. Of course, these noise levels depend
on the type of building, its orientation to the transmission line, and
the location of adjacent buildings that could block or reflect the
noise. Generally, therefore, sleep interference might occur in resi-
dences near a right-of-way.
Speech interference, or speech masking, may become a
problem at noise levels of around 55 dB(A). Because noise levels inside
VI-72
-------
buildings are about 10 to 30 dB lower than outdoor noise levels the same
distance from the lines, indoor noise generally will not cause speech
masking. However, out of doors, near the edge of a right-of-way, speech
masking might be expected during wet weather.
Noise experts disagree on the nature of the annoyance:
some say that annoyance is caused by speech or sleep interference; some
say it is a psychological reaction to noise separate from the other
effects. In either case, annoyance can be more dependent on the
characteristics of the noise and the meaning the noise conveys to the
person than on the actual noise level.
Like audible noise, radio and television interference are
caused by corona discharges on transmission lines and are therefore a
problem only in wet weather. The interference increases with higher
voltage lines and decreases with greater distance from the lines.
lonization of the air surrounding conductors during
corona discharge creates free oxygen radicals (0 ) that combine with
oxygen molecules (CO to produce ozone (0_). Ozone production dur-
ing wet weather is about 30 times the rate during fair weather. Calcu-
lations of the production of ozone from some proposed 765 kV lines in
New York state indicated that under worst conditions (wet weather and
slow wind moving parallel to a long straight stretch of lines for sever-
al hours), the lines would increase the ambient concentration of ozone
by 5 to 9 parts per billion (ppb) measured directly under the lines.
This increase in concentration would fall quickly as measurement points
moved laterally away from the lines. The present federal air quality
standard for ozone is 80 ppb not to be exceeded more than 1 hour per
year. Concentrations of 50 ppb are necessary before even the most
sensitive plants are damaged. Thus, ozone is not considered to be a
serious impact of transmission line operation.
Electrical Fields and Related Impacts—An operating transmission
line creates an electric field that is strongest at the surface of the
VI-73
-------
conductor, but decreases rapidly with increasing distance from the
conductor. Electric fields can cause shocks, may adversely affect the
operation of cardiac pacemakers, and are suspected by some people of
having physiological and psychological effects on animals and humans.
When an ungrounded object such as a bus is parked under a
transmission line operating at high voltage, the electric field induces
a voltage on the object. A grounded person (e.g., one standing on dirt
or in a puddle of water) touching the bus will create a path for current
to flow to the ground and will discharge the induced voltage. The dis-
charge will be felt as an electric shock similar to the electric shock
one experiences by walking across a carpet and touching a doorknob on a
dry day. The magnitude of the shock depends on the field strength
around the ungrounded object, the size and grounding of the object, the
strength of contact and grounding of the person, and the orientation of
the object in the field. The 765-kV lines proposed for New York state
were calculated to have the potential to be hazardous to a small,
grounded child touching a large, ungrounded vehicle in worst case
circumstances. The shock currents calculated for these lines were much
higher (by a factor of about 100) than shock currents measured under
experimental conditions. Therefore, electric shocks from transmission
lines should be considered a remote but real hazard.
Certain types of pacemakers may be affected by exposure to electro-
magnetic fields of the magnitudes found under 765-kV transmission
lines. Some pacemakers are designed to stand by while the heart is
functioning normally and begin pacing the heart when they pick up abnor-
mal electric signals from the heart. Because this type of pacemaker
must be sensitive to the electrical characteristics of the heart, arti-
ficial electrical interference caused by the patient standing in a large
electric field (from a transmission line, a microwave oven, or any of
several other sources) may cause the pacemaker to switch into its active
mode and begin giving signals to the heart. In that case, the heart
would be getting two conflicting signals — one natural and one from the
pacemaker. Interference of this type, while not optimal for the
VI-74
-------
patient, does not appear to be a serious hazard. Although many cases of
interference (none from transmission line operation) have been recorded,
no cases of death or serious damage from this type of interference have
been documented. Transmission lines operating at voltages less than 765
kV are not thought to cause pacemaker interference except in very un-
usual circumstances.
That electric fields produced by transmission lines
constitute a direct health hazard to people and animals has not yet been
proven to the satisfaction of the U.S. scientific community. Russian
experiments with substation workers report subjective neurologic dis-
orders such as headache, sluggishness, fatigue, and sleepiness. How-
ever, numerous experiments in the United States and other countries have
failed to duplicate these results. As with the other effects related to
the electrical characteristics of the lines, electric fields increase
with higher voltage lines.
Table VI-28
60 Hz ELECTRIC AND MAGNETIC FIELDS
Location of Measurement
Center of kitchen
30 cm (12 in) from
oven broiler
Private dwelling (away
from appliances)
Close to electric can
opener,electric shaver,
or hair dryer
Maximum field under
operating 765-kV lines
765-kV at 38 m (125 ft)
(edge of right-of-way)
765-kV at 152 m (500 ft)
Electric Field
(volts/meter)
130
10,000
2,500
100
Magnetic Field
(gauss)
0.001-0.1
5 or greater
0.5
0.15
0.01
VI-75
-------
Exposure to electric and magnetic fields is a relatively
common experience because all electric appliances create fields. Fields
from appliances differ somewhat from those of transmission lines in both
magnitude (see Table VI-28) and decreases of magnitude with distance
from the source.
b. Physical Impacts
Apart from their electrical characteristics, transmission
lines may affect an area simply by the physical presence of towers and
rights-of-way. A transmission line may disrupt the visual quality of an
area, or it may limit or affect the uses of the surrounding lands.
Visual Impacts—One of the most obvious effects of trans-
mission lines is the effect on the visual character of the area. The
severity of this effect can range from negligible in highly developed
areas with many existing utility lines to significant in relatively
undeveloped areas valued for their scenery. In flat terrain and in
areas characterized by low growing vegetation, the towers constitute the
major visual impact. Long, straight sections of transmission lines
present a definite visual disruption. In mountainous terrain, especial-
ly heavily vegetated areas, the right-of-way is visually disrupting; the
long strip of cleared land can be seen at greater distance than the
towers. Although the specific tower height and right-of-way width vary
under different circumstances, higher operating voltages require higher
towers and wider rights-of-way. Although good tower design and careful
siting can reduce the visual impact of transmission line corridors, more
intense visual disruption is to be expected as operating voltages in-
crease in the future.
Land Use Impact—Transmission lines can limit uses of the
surrounding land, and prohibit most uses of the right-of-way corridor
itself because of the necessity to prevent undue hazards or provide
access for surveillance and maintenance of the lines. Homes and air-
VI-76
-------
ports are notable examples of land uses that are limited around trans-
mission lines. In urban areas, the right-of-way may need to be fenced
to prevent access to the towers (which could constitute an "attractive
nuisance"). Like a fenced highway, a fenced utility corridor creates a
barrier to movement. Because of the shock hazard, buses should not stop
for passengers under or adjacent to the lines. Gasoline should not be
pumped into tanks where the fields are strong enough to produce a spark
between the pump nozzle and the vehicle.
The presence and operation of transmission lines may affect agri-
culture in and around the corridor in several ways. The towers and
lines may present a physical barrier to row cropping and to aerial
seeding, fertilizing, and spraying. Electrical fields from operating
transmission lines can induce voltages on large agricultural vehicles,
and cause shocks to farmers working with vehicles near the lines.
Similarly, metal irrigation equipment, especially the high sprayer type,
may be prohibited within a certain distance of the lines.
L. Gas Furnace
The only item of residential heating and cooling equipment that
does not have negligible environmental impact is the gas furnace in
System 1. The emission of air pollutants from a residential gas furnace
with an output of 70 MJ/hr (66,000 Btu/hr) were calculated from standard
emission factors. Those emissions are shown in Table VI-29.
M. 100-kW Fuel-Cell Power Plant
Emission characteristics of the 100-kW power plant were estimated,
based on rated-load operation. Under normal operation, the only
effluent stream leaving the power plant is the combined air exhaust and
reformer furnace flue gas. The composition of this stream (after
discharge from condenser E-5) is shown in Table VI-30. This stream
leaves the system saturated with water vapor at 127°C. Consequently,
fogging occurs on cold days.
VI-77
-------
Table VI-29
EMISSION OF AIR POLLUTANTS FROM A 70 MJ/HR
RESIDENTIAL GAS FURNACE
Pollutant
Emissions (g/hr)
NO
x
so2
CO
Particulates
Hydrocarbons
4.0
0.03
1.0
0.50
0.40
Source: Reference 2
Component
Table VI-30
COMPOSITION OF EFFLUENT STREAM
FROM 100-kW FUELr-CELL POWER PLANT
co2
HO (vapor)
N2
CO
CH4
NO
x
Particulates
Emission Rate at Rated Load, kg/hr
54.6
55.5
54.0
443.0
Nil*
*
Nil
See Table VI-28
Nil
Based on equilibrium combustion calculations
VI-78
-------
We have attempted to predict NO emissions, based on the
X
nonoptimized system design. Production of NO is expected during
operation of the reformer furnace burner. The method used was developed
by H. Shaw and begins with the calculation of equilibrium NO mole
21
fraction. The NASA-Lewis Equilibrium Program was used for this
purpose. The NO mole fraction in the furnace effluent was then
X
calculated for residence times of 5, 10, and 100 msec, based on the
equilibrium mole fractions of NO, OH, 0 , and N_, at the adiabatic
flame temperature and atmospheric pressure. As shown in Table VI-31,
NO emissions were found to be linear functions of residence time and
can thus be easily adjusted, if the residence time estimate is defined
more precisely. The method of Shaw is based on the Zeldovich mechanism
(including OH) with an empirical correction for "prompt NO ." Because
"prompt NO " would not be expected to contribute significantly at the
X
stoichiometry and for the fuel composition used in this study, we
neglected it. Also, no NO contribution from fuel-bound nitrogen
X
would be expected because no nitrogen compounds were specified in the
input fuel composition. The emission rate given in Table VI-31 was
calculated assuming that NO rather than NO is the pollutant.
Table VI-31
PREDICTED REFORMER FURNACE NOX PRODUCTION
Residence Time,a (msec)
5
10
100
Predicted N0y Levelsb
Mole Fraction Equivalent (gNOy/hr)
3.6 x 10
7.2 x 10
7.2 x 10
-7
-7
-6
831
1660
16,600
aAdiabtic flame temperature = 1642°C (2987°F.)
rated load, assuming NOX molecular wt. =30
VI-79
-------
As indicated earlier, the reformer furnace effluent temperature in
the nonoptimized base-case design was excessive. • In turn, that results
in relatively high levels of predicted NO emissions, even for short
X
residence times. Further study is required to develop a system inte-
gration scheme with lower furnace combustion temperature and lower N0x
emissions.
Although the environmental characteristics of fuel-cell power
plants are expected to be very favorable, little effort has been direct-
ed toward defining actual emission rates from complete power plants.
The only published data, reported by United Technology Corporation, and
quoted frequently, were obtained in 1970. Those data, shown in Table
VI-32, were probably measured using small natural gas-fired 12.5-kW fuel
cells developed by UTC for the TARGET program.
The NO rates shown are about 100 times lower than those predict-
x r
ed for the base-case system, indicating that proper operation of the
reformer furnace should not be a problem. New studies to measure the
pollution characteristics of phosphoric acid fuel cell power plants are
under way. The Department of Energy has included this technology in its
20
recent Environmental Development Plan.
Table VI-32
PUBLISHED POLLUTION CHARACTERISTICS OF EXPERIMENTAL
PHOSPHORIC ACID FUEL CELLS
Emission Type Emission Rate Range
(kg/MWh)*
S02 0 - 0.00014
N02 0.063 - 0.11
Hydrocarbons 0.014 - 0.10
Particulates 0 - 0.000014
^Quoted by UTC (22), based on a study by York Research Corporation,
Y-7306, April 1970. Fuel type was not specified.
VI-80
-------
Other pollution aspects of the 100 kW power plant are reviewed
below.
1. Spent Fuel-Cell Stack Disposal
The nominal life of phosphoric acid fuel-cell stacks is
assumed to be 40,000 hours. Thus, periodic stack replacement is re-
quired during the operating life of the power plant. For economic
reasons, stack disposal involves recovery and recycle of expensive
platinum electrocatalysts. This recovery step would take place at a
central facility, probably at the fuel-cell assembly plant site, or at a
catalyst supplier plant. The recovery process, as yet undefined, may
result in pollutant emissions.
Dismantling of the stack in an environmentally acceptable manner
will require procedures for disposal of the cell hardware (mostly
carbon/graphite components), probably via incineration. The 100 kW
power plant also contains about 15.3 kg of phosphoric acid electrolyte
and 61 kg of silicon carbide matrix. Suitable disposal procedures for
these potential pollutants must be developed.
2. Spent Catalyst Bed Disposal
The 100 kW power plant contains several catalyst beds that
require periodic replacement. These include the zinc oxide guard
chambers for odorant sulfur compound removal, the reformer reactor
(nickel catalyst), and the shift conversion reactors (copper and zinc
catalysts). Spent catalysts could be returned to the vendors for safe
disposal.
VI-81
-------
3. Normal Power Plant Operation
The principal gaseous emissions from the power plant were
discussed above. Other potential environmental impacts during normal
power plant operations include:
Possible phosphoric acid vapor release into the process air
stream, which limits fuel-cell life. Suitable guard beds,
probably containing iron adsorbents, used to prevent entry of
corrosive acid vapors into the water recovery condenser
systems, require periodic disposal.
Similar constraints apply to the use of circulating silicon
oil coolant, used to remove waste heat from the fuel cell.
New total energy power plant designs have switched to
water/steam or circulating air coolants.
The 100 kW power plant was designed to be water-conservative.
Thus, no impact on local water supplies is expected.
Noise pollution from the power plant has not been defined.
The plant contains rotating machinery equipment, such as fans,
blowers and pumps, with characteristic noise signatures.
Experimental data on prototype models are required to assess
noise levels.
VI-82
-------
N. References—Chapter VI
1. PEDCo. Environmental Specialists, Inc., "Wyoming Air Quality
Maintenance Area Analysis," U.S. Environmental Protection Agency,
Region VIII (May 1976).
2. "Compilation of Air Pollutant Emission Factors," U.S. Environmental
Protection Agency, AP-42 (February 1976).
3. A. J. Dvorak, et al., "The Environmental Effects of Using Coal for
Generating Electricity," Nuclear Regulatory Commission Report
NUREG-0252 (June 1977).
4. E. Goldman and P. J. Keller, "Water Reuse in the Electric
Generating Industry," National Conference on Complete Water Reuse,
1973.
5. "Development Document for Proposed Effluent Limitations Guidelines
and New Source Performance Standards for the Steam Electric Power
Generating Point Source Category," U.S. Environmental Protection
Agency, EPA 440/1-73/029 (1974).
6. G. D. Case, et al., "Health Effects and Related Standards for
Fossil Fuel and Geothermal Power Plants," Lawrence Berkeley
Laboratory Report No. LBL-5287 (January 1977).
7- P. Zubovic, "Geochemistry of Trace Elements in Coal," U.S.
Geological Survey (1975).
8. Radian Corporation, "Coal-Fired Power Plant Trace Element Study,"
U.S. Environmental Protection Agency, Region VIII (September 1975).
9. W. F. Holland, et al., "The Environmental Effects of Trace Elements
in the Pond Disposal of Ash and Flue Gas Desulfurization Sludge,"
Electric Power Research Institute (September 1975).
10. S. T. Cuffe and R. W. Gersite, "Emissions from Coal-Fired Power
Plants: A Comprehensive Summary," U.S. Department of Health,
Education and Welfare, Public Health Service (January 1970).
11. "Western Gasification Company Coal Gasification Project and
Expansion of Navajo Mine by Utah International, Inc., New Mexico —
Final EIS," U.S. Department of the Interior, Bureau of Reclamation
(January 1976).
12. Cameron Engineers, Inc., "Evaluation of Background Data Relating to
New Source Performance Standards for Lurgi Gasification," U.S.
Environmental Protection Agency Report EPA 600/7-77-057.
13. U.S. Environmental Protection Agency," Standards Support and
Environmental Impact Statement: Recommended Standards of
Performance for Coal Gasification Plants" (1976).
VI-83
-------
14. R. P. Hangebrauck, D. J. von Lehmden, and J. E. Meeker, "Emissions
of Polynuclear Aromatic Hydrocarbons and Other Pollutants from Heat
Generation and Incineration Processes," J. Air Pollution Control
Assn., 14(7): 267-78 (1964).
15. A. Attari, "Fate of Trace Constituents of Coal During Gasifi-
cation," U.S. Environmental Protection Agency Report
EPA-650/2-73-004 (August 1973).
16. U.S. Government Interagency Commercial Vehicle Post-1980 Goals
Study (1973).
17. Report of the Administrator of the U.S. Environmental Protection
Agency to the President and Congress on Noise (February 1972).
18. D. T. Beecher, et al., "Energy Conversion Alternatives Study —
Combined Gas/Steam Turbine Plant Using Coal-Derived Liquid Fuel,"
NASA CR-134942 (November 1976).
19. C. E. Jahnig, "Evaluation of Pollution Control in Fossil Fuel
Conversion Processes, Liquefaction: Section 2, SRC Process," U.S.
Environmental Protection Agency Report EPA-650/2-74-009-4 (1975)
and W. D. Shults, ed., "Preliminary Results: Chemical and
Biological Examination of Coal-Derived Liquids," Oak Ridge National
Laboratory Report ORNL/NSF/EATC-18 (1976).
20. U.S. Department of Energy, Environmental Development Plan,
Conservation Research and Technology, FY 1977, DOE/EDP-0017 (March
1978).
21. H. Shaw, "The Effects of Water, Pressure, and Equivalence Ratio on
Nitric Oxide Production in Gas Turbines," J. of Engineering for
Power, pp. 240-246 (July 1974).
22. United Technologies Corp., "Venture Analysis Case Study for On-Site
Fuel Cell Energy Systems," Vol. II, Final Report, p. A-ll (July 31,
1978).
VI-84
-------
VII. CAPITAL AND OPERATING COSTS OF SYSTEM COMPONENTS
This chapter describes the costs of constructing and operating the
components of the five energy supply systems described in Chapter IV.
The objective is to compare the economics of the five systems in terms
of the cost of heating and cooling residences and the system's capital
intensiveness. Sensitivities of the component costs to variations in
the costs of systems components are also analyzed.
Because comparison of the systems is a major objective of the
analysis, it is important to use consistent cost estimation procedures
for the various system components, insofar as possible. Thus, we
treated all system components as if they were operated by a regulated
utility, with the following exceptions: coal mine, liquid fuel
delivery, and residential heating and cooling equipment. This assump-
tion, which allows for a uniform cost analysis, is reasonable because
all major system components are associated with electrical generating
plants and the fuel networks that supply them.
The bases of the regulated utility economics used in the cost
analyses of the system components are as follows:
o All costs are in constant mid-1977 dollars.
o Plant sizes are set to take maximum reasonable advantage of
economies of scale and are characteristic of a mature industry.
o Source of capital is 35% equity (at a 15% rate of return) and
65% debt (at 10% interest).
o Plant construction, start-up, and operating schedule is shown
below. (The construction schedule would be shorter for
fuel-cell power plants and combined-cycle power plants.):
VII-1
-------
Depreciable Working
Year Investment Land Capital Expenses Revenue
-100%
1
2
3
4
5
6
Final
-5%
-20
-50
-25
—
—
—
-100%
+100
-60%
-100
-100
+60%
+100
+100
o
o
Project tax life is 20 years for coal conversion facilities,
fuel-cell power plants and combined-cycle power plants, and 30
years for pipelines and coal-fired power plants. Straight
line depreciation is assumed.
Plant investment factors:
- Start-up expenses, 5% of plant facilities investment (PFl)
- Working capital
Feed to liquid plants
Feed to coal plants
Labor for plants
Other cash expenses
Plant on-stream factors:
- Coal conversion facilities
- Intermediate load electrical
generating plants
- Pipelines
Labor factors:
- Operating labor (OL)
- Supervision (S)
- Maintenance labor (ML)
- Administrative & support labor
- Payroll burden, all labor
Maintenance supplies
Administrative expense
Property taxes and insurance
15 days
30 days
3 months
1 month
90%
35%
95%
$7.00/hour
15% OL
2% PFI
20% (OL+ML+S)
35%
2% PFI
2% PFI
2.5% PFI
VII-2
-------
o By-product credits will be assessed depending on plant loca-
tion and potential by-product markets.
o Land cost is $25,000 per hectare or $10,000 per acre (includes
basic site preparation)
o Income taxes include the 48% federal corporate income tax plus
4% state income tax for a total of 52%, applied against tax-
able income.
o Prices for energy supplied will be 20- or 30-year average
values.
o When appropriate, various contingency factors are applied to
the cost of an individual plant or facility depending on its
commercial experience. However, no across-the-board contin-
gency is applied.
o The utility rate base is the sum of depreciable investment
(PFI + interest during construction + start-up costs), land,
and working capital. Return on rate base (P) is defined as:
P = d(i) + (l-d)r,
where d = debt fraction,
i = interest rate on debt
1-d = equity fraction
r = return on equity.
An SRI computer program was used to calculate the 20- or 30-year
average cost of energy, based on the capital and operating factors
presented above.
The exceptions to these cost bases in the calculation of capital
and operating costs of system components will be addressed individually
in the following sections.
A. Coal Mine
The capital and operating costs of a 4.5 million tonne
(5 million ton) per year surface mine operating in the Powder River
Basin were evaluated using a cost model developed at SRI. The model
uses a simplified set of equations that relate mining costs to the
VII-3
-------
geological characteristics of the region (e.g., overburden and seam
thickness), size of the mine, and type of mine (area, open pit, or
contour). To calculate the mining costs, financial parameters must be
specified. The parameters used in the cost analysis are as follows:
o Mine life — 30 years
o Debt-to-equity ratio — 50/50
o Interest rate on debt — 10%
o Rate of return on equity — 15% discounted cash flow (DCF)
o Straight line depreciation.
Because various items of mining equipment must be replaced after 5,
10, or 20 years, they are depreciated on a separate schedule. Their
replacement costs are calculated as a deferred investment in the year of
replacement. These costs are incorporated into a financial model which
computes the coal mine revenue required to achieve the specified rate of
return on equity. The resulting cost of producing coal is an average
over the life of the mine. Using the mine characteristics discussed in
Chapter IV, the resulting capital investment required is shown in Table
VII-1.
The cost of producing coal from the mine is shown in Table VII-2.
Production costs include labor, supplies, union welfare payments, gen-
eral and administrative expense, and a reclamation fee mandated in the
Surface Coal Mining and Reclamation Act of 1977. Capital-related
charges include depreciation, depletion allowance, return on investment,
and debt payment. In addition, a $1.10/tonne ($1.00/ton) royalty pay-
ment for federal coal lease is assumed, along with state severence taxes
of 20% of the selling price (an approximate average of current Montana
and Wyoming state severence taxes). The resulting selling price of the
coal is $6.69/tonne ($6.08/ton).
VI1-4
-------
Table VII-1
CAPITAL INVESTMENT REQUIRED FOR A 4.5 MILLION TONNE
(5 MILLION TON) PER YEAR SURFACE COAL MINE
IN THE POWDER RIVER BASIN
Investment
($ Million) Percent
Primary Equipment 10.6 28
Overburden drilling 0.43
Overburden excavation 4.7
Coal drilling 0.06
Coal loading 1.09
Coal hauling 3.7
Spoil handling 0.22
Topsoil handling 0.37
Supporting Equipment 3.4 9
Total Equipment 14.0 37
Exploration 0.06
Land 0.20 1
Preproduction 13.2 35
Interest During Construction 2.7 7
Working Capital 7.9 20
Total Initial Investment 38.1 100
Deferred Capital Investment 17.5
VII-5
-------
Table VII-2
OPERATING COSTS FOR A 4.5 MILLION TONNE
(5 MILLION TON) PER YEAR SURFACE COAL MINE
IN THE POWDER RIVER BASIN
$ Million/year $/tonne ($/ton)
Labor 2.2 0.48 (0.44)
Operating crews 0.68
Hourly support 0.31
Salaried labor 0.46
Payroll burden 0.73
Welfare 4.3 0.95 (0.86)
Supplies 2.5 0.55 (0.50)
G&A 1.6 0.35 (0.32)
Taxes and Insurance 0.26 0.05 (0.05)
Reclamation Fee 1.75 0.39 (0.35)
Miscellaneous 0.30 0.06 (0.06)
Capital Charges 7.4 1.63 (1.48)
(debt payment, depreciation
return on investment, and
income taxes)
Royalties 5.0 1.10 (1.00)
Subtotal 25.3 5.57 (5.06)
Severence tax (20%) 5.1 1.12 (1.Q2)
30.4 6.69 (6.08)
VII-6
-------
B. Unit Train
The cost of unit train service is currently a controversial topic
because of the incipient competition from coal slurry pipelines.
Although some current unit train prices may be more affected by competi-
tion than by cost, the cost of shipping coal will be calculated, using
the financial parameters described at the beginning of the chapter. Two
factors that are important to unit train cost are:
o Existence of track of adequate quality along desired route
o Other rail traffic that can share track costs.
Estimated investment costs for 80 km (50 mi) of new single track
with siding and two unit trains are given in Table VII-3. Operating
costs are shown in Table VII-4, and are based on the assumption that the
80 km of new track is used to transport 23 million tonnes (25 million
tons) of coal per year, so that the unit train operation in question
must bear only 5.6% of the cost of constructing and maintaining the
track. The cost of shipping coal is $0.27 per GJ ($0.29 per million
Btu), or 0.44 cents per tonne-km (0.64 cents per ton-mile).
The cost of shipping coal is sensitive to the amount of traffic
sharing the 80 km of new track, as is evident in Figure VII-1. Coal
shipping costs range from 0.42 cents per tonne-km (0.62 cents per
ton-mile), assuming 45 million tonnes (50 million tons) per year of
traffic, to 0.89 cents per tonne-km (1.3 cents per ton-mile) assuming
that no other traffic uses the track.
C. Coal-Fired Power Plant
Table VII-5 gives a detailed estimate of the capital investment for
a new 800-MW plant along with the percentage of the cost associated with
the various items of equipment. The total capital investment is $674
million, or $843 per kW of installed capacity. Table VII-6 displays the
operating costs and revenue required for such a plant.
VII-7
-------
Table VII-3
ESTIMATED INVESTMENT FOR 80 km (50 mi) OF
NEW TRACK AND TWO 100-CAR UNIT TRAINS
Investment
Cost Component ($ Million) Percent
Rolling stock (2 unit trains)
8 locomotives, 3,000 hp each 2.4 23
200 aluminum-sided 91 tonne (100 ton)
(net) hopper cars 8.2 77
Subtotal for Trains 10.6 100
Track construction
Land acquisition cost 2.0 4
Grading and preparation 16.0 35
Structures and culverts 5.5 12
Roadway 4.5 10
Communication 2.0 4
Grade crossings 0.5 1
Ties 10.0 22
Rails 5.5 12
Subtotal for Track Construction 46.0 100
Interest during construction of track 4.9
Working capital 0.2
Organization and start-up expenses 2.4
Total Capital Investment 64.1
VII-8
-------
Table VII-4
OPERATING COST AND REVENUE REQUIRED FOR
NEW UNIT TRAIN, 80 km (50 mi) NEW TRACK,
AND 1,200 km (750 mi) EXISTING TRACK
Operating Cost
Raw materials
Diesel fuel at $0.09/liter
($.35/gal)
Maintenance materials
Total Raw Materials
Labor (including payroll burden)
Operating and supervision
Maintenance
Administrative and support
Total Labor
Fixed costs
General and administrative
expenses
Property taxes and insurance
Plant depreciation
Total Fixed Costs
Total Operating Costs
Return on rate base
and income tax
Total Revenue Required
Source of Revenue
Delivered coal
$ Million/Year
Cents/GJa
(Cents/106 Btu)
0.9
0.7
1.6
1.0
1.5
0.5
3.0
1.0
0.3
0.7
1.2
5.8
1.4
7.2
7.2
4 (4)
3 (3)
6 (6)
4 (4)
6 (6)
_2 (2)
11 (12)
4 (4)
1 (1)
2 (2)
~5 (5)
22 (23)
5 (5)
27.5 (29)
27.5 (29)
Assumes 80 km of new track is shared by 23 x 10^ tonnes per year
of coal traffic.
Totals may not equal sum of numbers in column because of rounding
errors.
VII-9
-------
I
8
13
-------
Table VII-5
CAPITAL INVESTMENT FOR 800-MW COAL-FIRED POWER PLANT WITH FGD
Investment
Plant Section ($ Million) Percent
Steam generators 133 27
Turbine generators and
associated equipment 86 18
Coal handling equipment 7 1
Stack 5 1
Other mechanical equipment, heating, etc. 12 2
Piping 49 10
Controls and instrumentation 12 2
Electrical equipment 12 2
Electrical bulk materials 38_ _8_
Subtotal Power Facilities 354 71
Electrostatic precipitator and ash handing 66 14
FGD system 72 15
Subtotal 492 100
Engineering and home office service,
including fees 54
Total Plant Facilities Investment 546
Land 5
Interest during construction 90
Organization and start-up expenses 27
Working capital 6_
Total Capital Investment 674
VII-11
-------
Table VII-6
OPERATING COSTS AND REVENUE REQUIREMENTS
FOR 800-MW COAL-FIRED POWER PLANT WITH FGD (35% LOAD FACTOR)
$ Million/Year
Operating Cost
Raw materials
Coal at $12.30/tonne
($11.20/ton) delivered 16
Lime at $39/tonne
($35/ton) 1
Maintenance materials __4
Total Raw Materials 21
Sludge disposal at $10/tonne
($97ton) 2
Labor
Operating and supervision 2
Maintenance 4
Administrative, support, and burden 4
Total Labor 10
a
Fixed costs
Administrative expense 6
Property taxes and insurance 7
Depreciation _H
Total Fixed Costs 24
Mills/kWh
6.5
0.4
1.6
8.5
0.8
0.8
1.6
1.6
4.0
2.4
2.9
4.5
9.8
Total Annual Operating Costs
57
23.1
Return on rate base and income tax"
30
12.2
Revenue Required for Electricity at Busbar
87
35.3
Assumes plant is 50% depreciated after 15 years of base-load
operation, and is then reassigned to intermediate load service.
VII-12
-------
The cost of electricity shown here is based on current construction
costs and a capacity factor (35%) typical of a cycling plant. In addi-
tion, the initial capital cost of the plant is assumed to be 50%
depreciated after 15 years of baseload service before reassigning the
plant to intermediate load service in the 1990s. The remaining
investment value of the plant is then depreciated over 30 years of
intermediate load service, and the yearly capital recovery factor is
reduced by one-half to reflect the reduced rate base. Therefore, the
capital recovery cost per unit of electricity will not change compared
to that for a baseload plant, since the annual capital recovery charges
and the load factor have both been reduced by a factor of two.
Figure VII-2 shows the sensitivity of the cost of electricity to
the delivered cost of coal and the capital cost of the plant.
D. Coal Gasification Plant
Table VII-7 shows the investment cost for each section of the SNG
ft ^ fi
manufacturing process for a plant making 7.8 x 10 nm (275 x 10 scf)
per day of SNG. Less than half of the required investment is for main
process plants; the majority (58%) of the investment is for support
facilities.
Operating costs and revenue required for the Hygas installation are
shown on Table VII-8. With a regulated utility rate basis, SNG is esti-
mated to be produced at a cost of $2.90 per GJ ($3.06 per million Btu).
Only ammonia is assumed to have a by-product value. If a large western
coal gasification industry were to develop, the remotely located by-
product sulfur is not expected to have a large market. Figure VII-3
shows the SNG price sensitivity to changes in coal price and in the
capital requirements of the process plants.
E. Coal Liquefaction Plant
Table VII-9 shows the details of the estimated capital investment
o
cost for an H-Coal plant producing 7,950 m per day (50,000 barrels
VII-13
-------
I
H-
•c-
I
0
UJ
fe
o
u
60
50
40
M 30
20
10 —
DELIVERED COAL COST - dollar per ton
10
15
10
DELIVERED COAL COST - dollar per tonne
I I
15
20
50
75 100 125
PLANT CAPITAL COST - percent of base case
150
175
FIGURE VII-2. SENSITIVITY OF THE COST OF ELECTRICITY TO PLANT CAPITAL
COST AND DELIVERED COAL COST
-------
Table VII-7
INVESTMENT REQUIRED FOR A 7.8 x 106 nm3 (275 x 106 scf)
PER DAY SNG PLANT BASED ON THE HYGAS PROCESS
Plant Section
Coal storage & reclaiming
Coal grinding
Coal slurry pumping
Gasification
Raw gas quench
Shift
Acid gas scrubbing
Methanation
Water reclamation
Sulfur recovery
Solids disposal
SNG drying
Steam & utilities
Water systems
Oxygen plant
General facilities
Contractor fees
Initial catalyst and chemicals
Investment
($ Million)
14
18
27
47
20
33
113
30
61
63
9
1
143
18
47
71
79
9
Percent
2
2
3
6
2
4
14
4
8
8
1
-
18
2
6
9
10
1
Total Plant Facilities Investment
803
100
Land
Interest during construction
Paid-up royalties
Working capital
Start-up costs
2
132
2
13
40
Total Capital Investment
992
VII-15
-------
Table VII-8
OPERATING COSTS AND REVENUE REQUIREMENTS FOR A 7.8 x 106 nm3
(275 x 10& scf) PER DAY SNG PLANT BASED ON THE HYGAS PROCESS
Operating Costs
Raw materials
Cents/GJ
$ Million/Year (Cents/10 Btu)
Coal at $6.69/tonne ($6.08/ton)
Water
Catalyst and chemicals
Maintenance materials
Total Raw Materials
38
1
4
16
64
44 (47)
1 (1)
5 (5)
19 (20)
69 (73)
Labor (including payroll burden)
Operating & supervision
Maintenance
Administrative and support
Total Labor
Fixed costs
Administrative expense
Property tax & insurance
Depreciation
Total Fixed Costs
Total Annual Operating Cost
Return on rate base & income tax3
Total Revenue Required
Sources of Revenue
SNG
By-product ammonia at $165/tonne
($150/ton)
3
16
4
4 (4)
18 (19)
5 (5)
23
255
256
4
255
27 (28)
16
20
49
85
167
88
19 (20)
23 (24)
56 (59)
98 (103)
194 (204)
101 (107)
295 (311)
290 (306)
_5 (51
295 (311)
20-year average values.
VII-16
-------
M
4.0
3.5
t 3.0
to
1
I
e>
CO
u. 2.5
O
LU
CO
5 2.0
UJ
CO
CO
1.5
1.0
50
I
COAL COST — dollar per ton
6 8
10
12
I
I
6810
COAL COST — dollar per tonne
I
I
12
75 100 125
PLANT CAPITAL COST - percent of base case
150
14
175
14
4.0
3.5
3
o
3.0 §
.o -a
I
O
z
CO
u.
O
2.0 ,_
CO
O
O
1.5
FIGURE VII-3. SENSITIVITY OF THE COST OF SNG TO PLANT CAPITAL COST AND COAL COST
-------
Table VII-9
CAPITAL INVESTMENT FOR A 7,950 m3 (50,000 bbl) PER DAY PLANT
PRODUCING DISTILLATE FUEL OIL FROM COAL BY THE H-COAL PROCESS
Investment
Plant Section ($ Million) Percent
Coal storage, handling, and preparation 45 5
Slurry preparation 17 2
Hydrogenation section 130 14
Product separation and fractionation 40 4
Steam reforming hydrogen 23 3
Partial oxidation hydrogen 170 19
Oxygen 50 5
Sulfur and ammonia recovery 85 9
Utilities and steam 180 20
General facilities 80 9
Contractor fees 9J) 10
Total Plant Facilities Investment 910 100
Land 2
Interest during construction 149
Paid-up royalties 2
Working capital 16
Start-up costs 46
Total Capital Investment 1,125
VII-18
-------
Table VII-10
CAPITAL INVESTMENT FOR A 7,630 m3 (48,000 bbl) PER DAY PLANT
PRODUCING NAPHTHA AND FUEL OIL FROM COAL BY THE H-COAL PROCESS
Investment
Plant Section ($ Million) Percent
Coal storage, handling, and preparation 45 5
Slurry preparation 17 2
Hydrogenation section 130 13
Product separation and fractionation 40 4
Steam reforming hydrogen 23 2
Partial oxidation hydrogen3 180 19
Oxygena 54 5
Sulfur and ammonia recovery3 87 9
Utilities3 190 20
General facilities3 84 9
Naphtha hydrotreater3 18 2
Contractor fees 95 10
Total Plant Facilities Investment 963 100
Land 2
Interest during construction 158
Paid-up royalties 2
Working capital 17
Start-up costs 48
Total Capital Investment 1,190
3Plant sections changed due to addition of naphtha hydrotreater.
VII-19
-------
per day) of distillate fuel oil. Table VII-10 gives the cost of the
same plant with the additional facilities necessary to hydrotreat the
naphtha portion of the product so that it is suitable for steam
reforming. The remainder of the distillate product (200-495°C or
400-925°F) is assumed to be sold as fuel oil. The added cost of
hydrotreating the naphtha includes the expanded oxygen plant, partial
oxidation gasifier, sulfur and ammonia recovery, and utilities, as well
as the cost of the hydrotreating plant. These new and expanded plant
sections add $65 million, or about 5.8%, to the base liquefaction plant
capital investment.
Tables VII-11 and VII-12 give the operating costs and revenue
requirements for fuel oil production without and including the naphtha
hydrotreating step. When producing only fuel oil the required revenue
is $3.02 per GJ ($3.19 per million Btu). If the naphtha is to be hydro-
treated for steam reforming, the non-naphtha distillate is assumed to be
sold as a by-product at $2.84 per GJ ($3.00 per million Btu). The cost
of hydrotreated naphtha is then $3.77 per GJ ($3.97 per million Btu).
Figure VII-4 shows the sensitivity of distillate fuel oil cost to
the cost of feed coal and to the capital cost of the process plant.
Figure VII-5 shows the sensitivity of hydrotreated naphtha to the cost
of feed coal and to the credit allowed for by-product fuel oil.
F. Gas Pipeline
The estimated investment required for an 81-cm (32-in.) gas pipe-
line is shown in Table VII-13. Cost for physical equipment and instal-
lation is $480 million, and total capital investment including equip-
ment, land, interest during construction, and working capital is
$548 million. Investment costs vary widely depending on construction
difficulty. The midwestern location for the pipeline in this study
should allow relatively easy construction because the route should not
pass through urban areas, mountainous areas, or very rocky soil. Table
VII-14 gives the operating costs and revenue required for the SNG
VII-20
-------
Table VII-11
OPERATING COSTS AND REVENUE REQUIREMENTS FOR A 7,950 m3
(50,000 bbl) PER DAY PLANT PRODUCING DISTILLATE FUEL OIL
FROM COAL BY THE H-COAL PROCESS
Operating Costs
Raw materials
Cents/GJ
$ Million/Year (Cents/10 Btu)
Coal at $6.69/tonne
Water
Catalyst and chemicals
Maintenance materials
Total Raw Materials
Labor (including payroll burden)
Operating & supervision
Maintenance
Administrative and support
Total Labor
Fixed costs
Administrative expense
Property tax and insurance
Depreciation
Total Fixed Costs
Total Annual Operating Cost
f%
Return on rate base and income tax
Total Revenue Required
Sources of Revenue
Distillate fuel oil
By-product ammonia at $165/tonne
20-year average values.
49
3
10
18
80
28
304
298
6
304
50 (53)
3 (3)
10 (11)
18 (19)
81 (86)
5
18
5
5 (5)
19 (20)
5 (5)
29 (30)
18
23
55
96
204
100
19 (20)
23 (24)
56 (59)
98 (103)
208 (219)
101 (107)
309 (326)
302 (319)
__7 (71
309 (326)
VII-21
-------
Table VII-12
OPERATING COSTS AND REVENUE REQUIREMENTS FOR A
7,630 m3 (48,000 bbl) PER DAY PLANT PRODUCING
NAPHTHA AND FUEL OIL FROM COAL BY THE H-COAL PROCESS
Operating Costs
Raw Materials
Cents/GJ
$ Million/Year (Cents/10 Btu)
Coal at $6.69/tonne
Water
Catalyst and chemicals
Maintenance materials
Total Raw Materials
Labor (including payroll burden)
Operating & supervision
Maintenance
Administrative and support
Total Labor
Fixed Costs
Administrative expense
Property tax and insurance
Depreciation
Total Fixed Costs
Total Annual Operating Cost
Return on rate base and income tax
Total Revenue Required
Sources of Revenue •
Hydrotreated naphtha at $3.77/GJ
($3.97/106 Btu)
By-product fuel oil at $2.84 GJ
($3.00/106 Btu)
By-product ammonia at $165/tonne
20-year average values.
49
3
12
Jl
83
29
166
147
7
320
51 (54)
3 (3)
12 (13)
20 (21)
"86
5
19
5
5 (5)
20 (21)
5 (5)
30 ( 31)
19
24
59
102
214
106
320
20
25
62
107
224
111
335
(21)
(27)
(65)
(113)
(237)
(117)
(354)
174 (184)
153 (162)
8 (8)
335 (354)
VII-22
-------
N>
UJ
D
CO
a
u.
o
fc
o
u
4.0
3.5
2.5
2.0
1.5
50
COAL COST - dollar per ton
6 8
10
12
I
I
14
6 8 10
COAL COST — dollar per tonne
I I
12
75 100 125 150
PLANT CAPITAL COST - percent of base case
14
4.0
3.5
£
c
o
3.0
a.
a
m
2.52
CO
2.0 a
fe
8
1.5
FIGURE VII-4. SENSITIVITY OF THE COST OF DISTILLATE FUEL OIL TO
PLANT CAPITAL COST AND COAL COST
-------
5.5
5.0
4.5
% 4.0
I
I
Q
UJ
3.5
CC
O
3.0
2.5
o
fc
o
o
2.0
1.5
1.0
I
2.0
COAL COST - dollar per ton
6810
12
I
I
I
I
2.5
I
6810
COAL COST - dollar per tonne
3.0
I
12
14
14
5.5
5.0 =
m
o
•D
4.0 |
I
Q_
•J C
3.5
Q
UJ
!c
3.0
DC
Q
2.5
2.0
1.5
3.5 dollar per million Btu
I
I I I
2.5 3.0 3.5
BYPRODUCT FUEL OIL CREDIT- dollar per GJ
FIGURE VII-5. SENSITIVITY OF THE COST OF HYDROTREATED NAPHTHA TO
PLANT CAPITAL COST AND BYPRODUCT FUEL OIL CREDIT
VII-24
-------
Table VII-13
CAPITAL INVESTMENT FOR A 81-cm (32-in.) DIAMETER GAS
TRANSMISSION PIPELINE — 1,300 km (800 mi)
Investment
Cost Component ($ Million) Percent
Line pipe 151 32
Pipe coatings 9.9 2
Valves 7.4 2
River and road crossings 0.2
Cathodic protection 0.2 —
11 compressor stations 107 22
Miscellaneous 9.2 2
Communications and metering 3.5 ^
Subtotal for Materials 296 62
Pipeline construction 116 24
Compressor station construction 27.2 6
Engineering design 37.3 8
Survey and mapping 2.5 1^
Subtotal for Services 183 38
Total for Construction of Pipeline 480 100
Land (right of way) and damages 6.4
Interest during construction 54.8
Working capital 7.5
Total Capital Investment 548
VII-25
-------
Table VII-14
OPERATING COSTS AND REVENUE REQUIREMENTS FOR AN 81-cm (32-in.)
DIAMETER GAS PIPELINE — 1,300 km (800 mi)
Operating Costs
Raw materials
Cents/GJa
$ Million/Year (Cents/10 Btu)
SNG fuel at $2.90/GJ
($3.06/million Btu)
Maintenance materials
Total Raw Materials
Labor (including payroll burden)
Operating and supervision
Maintenance
Administrative and support
Total Labor
Fixed costs
Administrative expense
Property tax and insurance
Depreciation
Total Fixed Costs
Total Annual Operating Costs
Return on rate base and income tax
Total Revenue Required
Source of Revenue
Delivered SNG
62
_4
66
10
12
17
39
110
48
158
158
26 (27)
_1 OJ
27 (28)
2
2
1
5
1
1
1
2
(1)
(1)
—
(2)
4 (4)
5 (5)
_2 O)
16 (17)
45 (47)
19 (20)
64 (68)
64 (68)
Totals may not equal sum of numbers in column because of rounding
errors.
VII-26
-------
pipeline. The $0.64 per GJ ($0.68 per million Btu) required revenue
assumes:
o Utility financing.
o $2.90 per GJ ($3.06 per million Btu) price for SNG at the
gasification plant.
o Thirty-year project and tax lives.
Figure VII-6 shows the sensitivity of required revenue to the
parameters of SNG price and pipeline capital cost.
Figure. VII-7 shows the sensitivity of gas transmission cost to
economies of scale. The costs for the 81-cm (32-in.) diameter pipeline
chosen as an example in this report could be significantly changed if a
different diameter pipeline were used. Moreover, existing pipelines
that are fully depreciated could offer transportation service at consid-
erably reduced prices. As shown on Table VII-14, the cost of delivering
SNG in a newly capitalized pipeline is $0.64 per GJ, but an older system
with no capital charges or income tax would have a cost of $0.36 per GJ.
If an existing system were also of larger diameter (say 122 cm), the de-
livered cost would drop to about $0.26 per GJ.
G. Liquids Pipeline
o
The 15,900 m (200,000 barrel) per day liquid pipeline is smaller
in diameter (51 cm) than the SNG pipeline (81 cm), so that the liquid
pipeline is less expensive than the SNG line. In addition, the liquid
pipeline uses pumping stations rather than more expensive compressor
stations. Estimated investment costs for the liquid pipelines are shown
in Table VII-15. Total capital investment is $309 million for a
1,300-km (800 mi) pipeline.
Table VII-16 shows the operating cost and revenue requirements for
fuel oil pipeline shipments. A cost is included for diesel oil to fuel
VII-27
-------
1.2
1.0
SNG COST - dollar per 106 Btu
2.0 3.0 4.0 5.0 6.0
7.0
T
1 r
1.2
1.0
13
&
,. 0.8
_n
1
I
0.6
cc 0.4
eo
BASE CASE
1.0
10
o
0.8
CO
O
0.6
CO
CO
CO
0.4 <
I-
CO
0.2
0.2
0.0
I
0.0
1.0
2.0 3.0 4.0 5.0
SNG COST - dollar per GJ
6.0 7.0
50
75 100 125
PIPELINE CAPITAL COST - percent of base case
150
FIGURE VII-6. SENSITIVITY OF SNG TRANSMISSION COST TO
PIPELINE CAPITAL COST AND SNG COST
VII-28
-------
4.0
8 12 16 20
I I
PIPELINE DIAMETER - in.
24 28 32 36
40 44 48
1 T
3.5
3.0
o
Z 2.5
O
E 2.0
tn
<
a
UJ
< 1.B
UJ
1.0
BASE CASE
0.5
I
I
20
40
60 80
PIPELINE DIAMETER - cm
100
120
FIGURE VII-7. EFFECT OF PIPE DIAMETER ON SNG TRANSMISSION COSTS
(1300 km Distance)
VII-29
-------
Table VII-15
CAPITAL INVESTMENT FOR A 51-cm (20-in.) DIAMETER
COAL LIQUIDS PIPELINE — 1,300 km (800 mi)
Cost Component
Line pipe
Pipe coating
Valves
Cathodic protection
Miscellaneous
Communication and metering
Subtotal for Materials
Pipline construction
10 pump stations construction
Engineering
Survey and mapping
Subtotal for Services
Total for Construction
of Pipeline
Land (right of way) and damages
Interest during construction
Working capital
Total Capital Investment
Investment
($ Million)
80.4
5.7
2.4
0.2
7.5
5.4
102
98.0
22.4
22.2
2.4
145
247
4.0
25.3
33.1
309
Percent
33
2
1
—
3
2
41
40
9
9
1
59
100
VII-30
-------
Table VII-16
OPERATING COSTS AND REVENUE REQUIREMENTS FOR A
51-cm (20-in.) DIAMETER LIQUIDS PIPELINE -- 1,300 km (800 mi)
Cents/GJ
$ Million/Year (Cents/106 Btu)
Operating Costs
Raw materials
Diesel fuel for pumps 6.9
Maintenance materials 0.2
Total Raw Materials 7.1
Labor (including payroll burden)
Operating and supervision 1.6
Maintenance 2.0
Administrative and support 0.7
Total Labor 4.3
Fixed costs
Administrative expense
Property tax and insurance
Depreciation
Total Fixed Costs
Total Annual Operating Costs 31.6
Return on rate base and income tax
Total Revenue Required 62.5
Source of Revenue
2 (2)
2 (2)
0.5 (0.5)
1 (1)
4.9
6.2
9.1
20.2
31.6
30.9
1
1.5
2
5
8
8
(1)
(1.5)
(2)
(5)
(8)
(8)
Fuel Oil
62.5
15 (16)
15 (16)
VII-31
-------
the pump drivers that must be purchased from a source external to the
pipeline operation. The total cost of shipping coal-derived distillate
fuel oil is $0.15 per GJ ($0.16 per million Btu).
When coal-derived naphtha and fuel oil are sent through the same
pipeline, the shipping costs change slightly because of the difference
in heating value between the two products. The cost of shipping naphtha
is $0.16 per GJ ($0.17 per million Btu), while that of the fuel oil is
$0.14 per GJ ($0.15 per million Btu).
As in the SNG pipeline case, a larger diameter liquid pipeline
would lower the naphtha and fuel oil shipping cost on a heating value
basis. Figure VII-8 shows the sensitivity of fuel oil shipment cost to
the capital cost of the pipeline.
H. Liquid Fuel Distribution
Important factors that determine the cost of transporting distil-
late fuel by train are shipping distance, car size, number of cars per
train, terrain, train speed, and loading and unloading times. Represen-
tative costs of transporting fuel by unit train are shown in Figure
VII-9. These costs are based on 37,850-liter (10,000-gallon) tank cars,
100 cars per train, and 56 km-per-hour (35 mph) average train speed.
Costs include standing time for loading and unloading and empty back-
haul. The annual capital charges are assumed to be 20% of the invest-
ment. The resulting capital charges account for 15% of the shipping
cost, with operating and maintenance costs making up the remaining 85%.
Costs of trucking naphtha are shown in Figure VII-10. A
34,000-liter (9,000-gallon) truck, operating at an average road speed of
64 km per hour (40 mi per hour), with a one-stop delivery and empty
backhaul, is assumed. Capital charges account for 40% of the trucking
cost based on a 30% annual charge rate, and operating and maintenance
costs make up the remaining 60% of trucking costs. Trucks with smaller
VII-32
-------
0.30
0.30
0.25
o 0.20
8
I °-15
w
CO
X
HI
0.10
O
o
BASE CASE
0.25
0.20
3
4->
CO
fc
a
O
O
0.15 55
CO
CO
0.05
0.10
0.05
cc
O
I
I
50
75 100 125
PIPELINE CAPITAL COST - percent of base case
150
FIGURE VII-8. SENSITIVITY OF LIQUID FUEL TRANSMISSION
COST TO PIPELINE CAPITAL COST
VII-33
-------
0.8
200
DELIVERY DISTANCE - miles
400 600
800
3
I
k_
JO
1
I
oc
2
en
<
tr.
<
cc
0.4
0.2
I
I
0.8
1
0.6 I
£
u
JO
o
0.4
0.2
£
8
o
a.
cc
200
400
600
800
1000 1200
1400
1600
DELIVERY DISTANCE - km
FIGURE VII-9. RAILROAD TANK CAR TRANSPORT COSTS (Fuel Oil)
VII-34
-------
capacity, lower operating speed, and making more deliveries will have
higher trucking costs than those indicated in Figure VII-10.
To estimate the cost of distributing distillate fuel via train to a
centralized combined-cycle power plant and of distributing naphtha to
dispersed 26-MW fuel-cell power plants requires that assumptions be made
about the relative distances of these facilities from the pipeline ter-
minus. A reasonable assumption is that the pipeline terminates near any
of the three cities under consideration — Omaha, Des Moines, or Kansas
City. Because fuel-cell power plants would be located near load
centers, the distribution of naphtha by truck would involve relatively
short distances. If an average distance of 40 km (25 mi) is assumed,
the cost of distributing naphtha would be about $0.07 per GJ ($0.07 per
million Btu). To this distribution cost must be added the cost of
storing the naphtha at the bulk storage terminal prior to delivery.
This cost has been estimated to be about $0.01 per GJ ($0.01 per million
Btu). Thus, the total storage and delivery cost of naphtha is about
$0.08 per GJ ($0.08 per million Btu).
If the combined-cycle power plant is assumed to be centrally
located with respect to the three cities, a distance from the bulk
terminal to the plant of about 160 km (100 mi) would be reasonable. The
total cost of shipping distillate fuel via railroad tank cars (including
storage costs) would then be approximately $0.12 per GJ ($0.12 per
million Btu).
Because of the low cost of shipping liquid fuels compared to their
production costs, it is clear that large changes in the assumed shipping
distances will not significantly affect the delivered cost of these
fuels.
I. Gas Distribution
The costs of distributing natural gas from large interstate pipe-
lines to individual customers vary widely around the country. A number
VII-35
-------
I
U>
0.4
0.3
I
[2 0.2
8
o
o
oc 0.1
25
50
DELIVERY DISTANCE - miles
50 75 100
I
100 150
DELIVERY DISTANCE - km
125
200
150
T
0.4
0.3
TJ
0.2 £
o
0.1 {£
250
FIGURE VII-10. TANK TRUCK TRANSPORTATION COSTS (Naphtha)
-------
of key variables, including density of the distribution network and age
of equipment, are important in determining these costs. Because of the
many possible variables, it is more appropriate to use actual costs of
gas distribution rather than to attempt to derive such costs from esti-
mates of capital and operating costs.
Historically, the cost of transmitting and distributing natural gas
has been a large fraction of the total cost of gas paid by residential
customers. From 1971 to 1975 the average wellhead price of natural gas
3
increased from 0.643 to 1.57 cents per nm (18.2 to 44.5 cents per
o
1,000 scf). During the same period the average residential price
3
increased from 4.06 to 6.11 cents per nm ($1.15 to $1.73 per
o
1,000 scf). The difference between the wellhead and selling price is
the cost of gas transmission and distribution; it increased from
3.43 to 4.52 cents per nm3 ($0.97 to $1.28 per 1,000 scf). Thus, gas
transmission and distribution costs are large and have been increasing,
although not nearly so rapidly as wellhead gas prices.
The cost of distributing gas, as opposed to the total cost of
transmission and distribution, can be derived from data given in the
2
American Gas Association publication, Gas Facts. The distribution
costs can be approximated as the difference between the price of gas
paid by local gas utilities to pipeline companies (gas sold for resale)
and the price by residential, commercial, or industrial customers. To
determine costs specific to the region of interest, we used data from
the West North Central states, which encompass Omaha, Des Moines, and
Kansas City.
In 1975 the average price paid for natural gas by gas utilities was
$0.66 per GJ ($0.70 per million Btu). The average prices paid by resi-
dential and commercial customers were $1.30 and $1.04 per GJ ($1.37 and
$1.10 per million Btu), respectively. Therefore, the cost of distribu-
ting gas to residential and commercial customers may be estimated at
$0.54 and $0.38 per GJ ($0.57 and $0.40 per million Btu). Assuming that
distribution costs increased at about the same rate from 1975 to 1977 as
VII-37
-------
they did from 1971 to 1975 (about 7% per year), the 1977 cost of dis-
tributing natural gas in the West North Central states would be $0.62
per GJ ($0.65 per million Btu) for residential customers and $0.44 per
GJ ($0.46 per million Btu) for commercial customers.
In subsequent cost calculations the cost of distributing natural
gas to dispersed 26-MW fuel-cell power plants will be assumed to corre-
spond most nearly to that of distributing gas to large commercial
customers.
J. Combined-Cycle Power Plants
The capital investment required for a 270-MW combined-cycle power
plant is shown in Table VII-17. The total capital cost of $86.1 million
represents an investment of $319 per kW of installed capacity. This
total includes the cost of fuel treatment for the coal derived distil-
late fuel oil and of advanced gas turbines.
The cost of generating electricity in intermediate load operation
(35% capacity factor) is shown in Table VII-18, based on a delivered
cost of H-Coal distillate of $3.29 per GJ ($3.47 per million Btu). The
high cost of this fuel results in fuel-related costs of nearly half the
cost of generating electricity. The total cost of electricity from the
combined-cycle plant is 46.9 mills per kWh.
Figure VII-11 displays the sensitivity of the cost of electricity
to the cost of distillate fuel and the plant capital investment.
K. 26-MW Fuel-Cell Power Plant (SNG)
Manufacturing costs for the molten carbonate fuel-cell power plant
were estimated using the bases discussed below. These bases apply to
both the System 2 and System 3 dispersed site power plants. The
VII-38
-------
Table VII-17
CAPITAL INVESTMENT FOR A 270-MW COMBINED-CYCLE POWER PLANT
USING DISTILLATE FUEL FROM THE H-COAL PROCESS
Investment
Plant Section ($ Million) Percent
Fuel preparation 2.4 3
Gas turbine-generator sets 20.0 27
Waste heat boilers 4.7 6
Steam turbine-generator set 15.0 20
Process mechanical equipment 5.9 8
Electrical equipment 6.5 9
Civil and structural 5.8 8
Piping and instrumentation 4.2 6
Engineering & home office services 6.8 9
Miscellaneous 3.1 4
Total Plant Facilities Investment 74.4 100
Land 0.3
Interest during construction 6.0
Organization and start-up expenses 3.9
Working capital 1.5
Total Capital Investment 86.1
VII-39
-------
Table VII-18
OPERATING COSTS AND REVENUE REQUIREMENTS FOR A
270-MW COMBINED-CYCLE POWER PLANT USING DISTILLATE
FUEL FROM COAL (35% LOAD FACTOR)
Operating Costs $ Million/Year Mills/kWh
Raw materials
H-Coal distillate @ $3.29 per GJ
($3.47 per million Btu)
delivered 19.6 24.2
Water 0.1 0.1
Maintenance materials 1.2 1.4
Labor
Operating and supervision 0.6 0.7
Maintenance 1.7 2.1
Administrative and support 0.5 0.6
Fixed costs
General administrative expenses 1.5 1.8
Property taxes and insurance 1.9 2.3
Depreciation 4.2 5.1
Return on rate base and income tax 7.5 9.1
Total Revenue Required 38.8 46.9
Source of Revenue
Electric Power 38.8 46.9
VII-40
-------
80
1.0
DISTILLATE FUEL COST - dollars per 106 Btu
2.0 3.0 4.0 5.0
6.0
7.0
T
T
T
70
60
1
fc 50
a
I
0 40
£
uj
LU
li.
° 30
20
10
I
I
1.0
2.0 3.0 4.0 5.0
DISTILLATE FUEL COST - dollars per GJ
I
I
6.0
7.0
50
75 100 125
POWER PLANT CAPITAL COST - percent of base case
150
FIGURE VII-11. SENSITIVITY OF COST OF ELECTRICITY TO POWER PLANT
CAPITAL COST AND DISTILLATE FUEL COST
VII-41
-------
following system components are included: fuel-cell trailer; reformer
module; equipment module; condenser (E-7); and power conditioner. Those
modular assemblies are designed to be transported to the plant site,
installed on concrete pads, and interconnected with preassembled sec-
tions of piping and ductwork.
1. Fuel-Cell Trailer Cost
The shippable fuel-cell trailer, rated at 3.34 MW, contains
eight stacks. Fabricated from structural steel, it is enclosed by
corrugated steel panels. The stack cost estimates are based on the
molten carbonate fuel-cell design described by UTC in the EGAS pro-
3
gram. Supplementary information was obtained from recent ERG
reports. The estimates were based on calculations of actual quan-
tities of raw materials used to fabricate the components. Current cost
of raw materials, in the form expected to be used, were obtained from
vendor contacts. Fabrication costs were then determined by multi-
plying the material cost by a manufacturing cost factor, which was
selected based on the production rate and the degree of automation
envisioned for the manufacturing facility. The factors reflect manu-
facturing value added, including direct and supervisory labor plus other
manufacturing burdens (e.g., maintenance and inventory costs).
The accuracy of this type of estimate mainly depends on the
assumptions made concerning the design configuration, such as material
thickness and the selection of the materials. The application of
incorrect manufacturing cost factors presents a lesser risk because they
tend to fall within a fairly predictable range for all manufacturing
facilities operating at high production rates. It can safely be assumed
that profitable businesses will employ the most advanced methods in
order to remain competitive.
The fuel-cell stack components and vendor quotes for materials
costs are listed in Table VII-19. The estimated cost of an individual
VII-42
-------
80
DISTILLATE FUEL COST - dollars per 106 Btu
1.0 2.0 3.0 4.0 5.0 6.0 7.0
~l I 1 1 1 1 T
70
60
I
E
I
ID
111
i
50
40
30
20
10
I
1.0
2.0 3.0 4.0 5.0
DISTILLATE FUEL COST - dollars per GJ
I
I
6.0
7.0
50
75 100 125
POWER PLANT CAPITAL COST - percent of base case
150
FIGURE VII-11. SENSITIVITY OF COST OF ELECTRICITY TO POWER PLANT
CAPITAL COST AND DISTILLATE FUEL COST
VII-41
-------
following system components are included: fuel-cell trailer; reformer
module; equipment module; condenser (E-7); and power conditioner. Those
modular assemblies are designed to be transported to the plant site,
installed on concrete pads, and interconnected with preassembled sec-
tions of piping and ductwork.
1. Fuel-Cell Trailer Cost
The shippable fuel-cell trailer, rated at 3.34 MW, contains
eight stacks. Fabricated from structural steel, it is enclosed by
corrugated steel panels. The stack cost estimates are based on the
molten carbonate fuel-cell design described by UTC in the EGAS pro-
3
gram. Supplementary information was obtained from recent ERG
reports. The estimates were based on calculations of actual quan-
tities of raw materials used to fabricate the components. Current cost
of raw materials, in the form expected to be used, were obtained from
vendor contacts. Fabrication costs were then determined by multi-
plying the material cost by a manufacturing cost factor, which was
selected based on the production rate and the degree of automation
envisioned for the manufacturing facility. The factors reflect manu-
facturing value added, including direct and supervisory labor plus other
manufacturing burdens (e.g., maintenance and inventory costs).
The accuracy of this type of estimate mainly depends on the
assumptions made concerning the design configuration, such as material
thickness and the selection of the materials. The application of
incorrect manufacturing cost factors presents a lesser risk because they
tend to fall within a fairly predictable range for all manufacturing
facilities operating at high production rates. It can safely be assumed
that profitable businesses will employ the most advanced methods in
order to remain competitive.
The fuel-cell stack components and vendor quotes for materials
costs are listed in Table VII-19. The estimated cost of an individual
VII-42
-------
00
Component
Table VII-19
FUEL-CELL STACK COMPONENTS
Dimensions, cm (in) Materials Weight/cell, kg (lb) Cost, $/kg ($/lb)
Source
Electrolyte 100.3 x 100.3 x 0.089 40% Li A102 0.89
tile (39.5 x 39.5 x 0.035) 21* Li2C03 0.92
39% K2C03 0.86
Total Tile Weight: 2.62
Anode
Cathode
Cathode
collector
Anode
collector
Cell
separator
92.2 x 92.2 x 0.076 Ni Powder 2.51
(36.3 x 36.3 x 0.030)
(56% Porosity)
92.2 x 92.2 x 0.038 Ni Powder 1.26
(36.3 x 36.3 x 0.015
(56% Porosity)
100.3 x 100.3 x 0.152 304 SS 0.75
(39.5 x 36.3 x 0.060)
Stock Thk - 0.010
(0.004)
92.2 x 92.2 x 0.102 304 SS 0.69
(36.3 x 36.3 x 0.040)
Stock Thk - 0.010
(0.004)
100.3 x 100.3 x 0.102 304 SS 1.14
(39.5 x 39.5 x 0.040)
Total Cell Weight: 9.02
Stack end 120.6 x 120.6 x 7.6 430 SS
plates (47.5 x 47.5 x 3)
Stock Thk - 0.952
(0.375)
Tie rods (16) 1.9 (0.75) Diam. 430 SS
Springs (16) — 302 SS
Total Stack Weight:
(1.95)
(2.03
(1.90)
(5.88)
(5.53)
(2.77)
(1.64)
(1.51)
(2.51)
(19.84)
6.16
1.94
0.51
2.88
352 (775)/stack 1.78
81 (178)/stack 1.78
109 (240)/stack 2.16
5,141 (11,311)
(2.80)
(0.88)
(0.23)
Lithium Co.
(America)
(1.31) (Average)
5.94 (2.70) International
Nickel
5.94 (2.70) International
Nickel
4.27 (1.94) Rodney Metals
4.27 (1.94) Rodney Metals
4.27 (1.94) Rodney Metals
(0.81) U.S. Steel
(0.81) U.S. Steel
(0.98) U.S. Steel
-------
fuel-cell trailer is $359,000, as shown in Table VII-20. The piping,
wiring, enclosure, and assembly material cost was arbitrarily based on
an assumed structure weight of 3,270 kg (7,200 Ib), costed at $2.20/kg
($l/lb). The assigned cost factor of 3.0 reflects a less automated,
more labor-intensive, manufacturing operation.
Table VII-20 shows that the costliest items are the fuel-cell
components, which make up 87.5% of the total. To minimize labor costs,
the most advanced production machinery would have to be used. The
capital investment for this facility has not been estimated, but it
would probably be quite high.
Table VII-20
FUEL-CELL TRAILER COST SUMMARY
Total
Raw Material Mfg. Cost Mfg. Cost Costb
% Total
Componenta Cost, $1000
Electrolyte
tile
Anode
Cathode
Collectors
Separators
End plates
Tie rods
Springs
Miscellaneous0
Total
31.4
60.9
30.5
24.9
19.9
5.0
1.2
1.9
7.2
182.9
Factor
1.2
0.6
0.6
1.2
1.2
1.3
0.6
0.6
3.0
$1000
37.7
36.6
18.3
29.9
23.8
6.5
0.7
1.1
21.6
176.2
$1000
69.1
97.5
48.8
54.8
43.7
11.5
1.9
3.0
28.8
359.1
Trailer Cost
19.2
27.2
13.6
15.3
12.2
3.2
0.5
0.8
8.0
100
Eight stacks/trailer.
b
Total cost = raw material + manufacturing cost.
c
Piping, wiring, enclosure, and assembly.
VII-44
-------
Considerable manufacturing development effort will be required
for the electrolyte tile production facility. A major problem area will
be tile cracking, unless a flexible tile configuration can be developed.
Electrolyte tiles are currently manufactured in a noncontinuous process.
Lithium aluminate powder is mixed with finely ground lithium and potas-
sium carbonate. The mixture is then placed in a mold, compressed, and
fired in a furnace. Ultimately, a completely automated production
facility should be used, similar to those developed for electrode manu-
facture. At present, sintered nickel electrodes are manufactured
commercially in a continuous 30-cm (12-in.) wide strip using a slurry
method for applying the nickel powder to a nickel-plated steel sub-
strate. Material cost at present is 50% of the total manufactured
cost. The manufacturer hopes to increase this to 75% with improved
methods.
Fabrication of collectors and separators will be fairly
straightforward. The collectors can be formed in large stamping
presses, which would require some handling of individual pieces. A
continuous roll forming production line could be used, which could
result in a cost factor lower than the assumed 1.2. The separator,
however, is a more complex component and will always require more
labor. It consists of an outer frame containing metal seal surfaces and
fuel manifolding, welded to the cell separating sheet. This con-
figuration will require handling of more than one part and seam welding
to join the parts together.
The assumed thicknesses of the electrolyte tile (0.089 cm),
collectors (0.010 cm), and separators (0.010 cm) are reasonably opti-
mistic projections, based on the current status of molten carbonate
fuel-cell design. The latter are fabricated from 304 stainless steel.
If corrosion of the stainless steel parts requires that they be replaced
by nickel, their material cost will almost double. Using nickel should
not affect the manufacturing cost, however, and would only increase the
total cost by 14%.
VII-45
-------
A comparatively small portion of the total cost (12.5%) is for
the stack hardware, structures, and enclosure. The production rates of
these items are too low to justify highly sophisticated and automated
machinery — thus a higher manufacturing cost factor was used.
Developing the manufacturing technology for fabricating most
components should not be difficult. The exception to this may be the
manufacture of the electrolyte tile. A more flexible electrolyte
structure is needed and perhaps can be developed.
2. Reformer Module Cost
The cost of the reformer module was calculated from Exxon data
on cylindrical-type reformer furnaces. The total material and labor
costs were determined for the fabrication of a single unit. A direct
labor rate of $9/hr was assumed and a factory overhead rate of 200% was
used. A learning factor of 0.9 was then applied to the labor cost to
determine the average cost per unit for 400 modules. Eighty percent of
the cost of the reformer is for the reactor and heat exchanger tubing
and manifolding. Stainless steel is required throughout because of the
high operating temperature. The total cost of the reformer is $504,000,
of which 43% is the cost of the heat exchangers (E-l through E-4).
3. Equipment Module Cost
The equipment module cost was determined by calculating the
total F.O.B. cost of heat exchangers, blowers, electric motor drivers,
and other components based on Exxon cost data and discussion with
vendors. The cost breakdown is shown in Table VII-21. The use of
canal-type recuperators for exchangers E-5 and E-6 resulted in substan-
tial cost reduction.
VII-46
-------
Table VII-21
EQUIPMENT MODULE COST BREAKDOWN
Item Cost, $1000
Exchanger E-5 9.6
E-6 58.2
ZnO guard bed 3.0
Knockout drum 1.5
Blower B-l 12.3
B-3 44.9
B-4 20.6
B-5 21.0
P-l 2.5
Module fabrication 55.0
Total Equipment Module Cost 228.6
4. Condenser Cost
The condenser section, exchanger E-7, consists of two bays of
air-fin heat exchangers, 3.4 x 9.2 m (11 x 30 ft). Each bay has two
fans. The cost of these units ($15,000 per bay) was determined from
Exxon cost data. Each bay would be shipped to the power plant site and
erected on concrete piers.
5. Power Conditioning Costs
Projected costs of power conditioning equipment have been
reported by Westinghouse. They vary from $50 to $70/kW, depending on
the input DC voltage. An average cost of $60/kW was assumed for this
s tudy.
VII-47
-------
6. Total Power Plant
Total installed costs were estimated for the base case
System 2 power plant. Total manufactured costs were calculated by
summing the costs for each system component. Installation-related costs
were then estimated. Here, site preparation costs include grading and
installation of access roadways, but not the cost of land. No cost
allowance was made for buildings and similar facilities, because the
system is assumed to operate unattended.
Other indirect costs were estimated as 40% of total manufac-
tured cost and 25% of total installation costs. These indirect costs
reflect general and administrative expense, taxes and insurance,
interest on investment, sales and marketing expense, return on invest-
ment, and contingencies, if any. Architect and engineering charges are
not explicitly detailed. Also, escalation and interest charges during
construction were not included. Site preparation and installation time
for the modular power plant is assumed to be short.
A breakdown of the estimated power plant cost is given in
Table VII-22. Total installed cost for the base-case system is
$12,530,000, equivalent to $522/kW of net output (24.0 MW). This value
is higher than expected, but substantial investment cost reduction is
possible.
The ultimate optimization of any power plant is a complex
trade-off between investment charges and fuel charges, reflecting
constraints placed on the system. The power plant here has been
constrained to meet a heat rate of 7,910 kJ/kWh (7,500 Btu/kWh) and to
be water-conservative. If those constraints were relaxed, investment
costs and the cost of delivered energy could be reduced.
For example, in many locations, the amount of water necessary
for reforming is readily available from local supplies and total water
conservation would not be necessary. Water requirements are about
VII-48
-------
Table VII-22
NOMINAL 26-MW FUEL-CELL POWER PLANT COST ESTIMATE
Item
No. in
System
8
Fuel-cell trailer
Reformer/heat exchanger
package
Equipment module
Condenser (E-7)
Power conditioner and
switchgear 4
Electrical wiring, controls
and instrumentation
Total Manufactured Cost
Site preparation
Freight and insurance
Mechanical structures,
foundations and piping
Total Installation Costs
Other indirect costs + profit3
Total Power Plant Installed Cost
Unit
Cost,
$1000
359.1
402
Total
Cost,
$1000
2,873
4
4
4
504
228.6
60
2,016
914
240
1,608
534
8,185
481
82
294
857
3,488
12,530
Taken as 40% of total manufactured cost + 25% of total installation costs,
VII-49
-------
0.57 liter/kWh (0.15 gal/kWh). In the tricity study region, 1 kWh is
worth 30-50 mills. Water consumption would only add 0.07 mills/kWh, a
negligible amount. As a result, E-2, the knockout drum, blower B-2, and
the water recycle pump could be eliminated, and the size of E-5 could be
decreased. Fuel cell performance and the heat exchange characteristics
of exchanger E-5 would also be affected; thus, although it is not clear
exactly what the final cost of electricity will be, the investment cost
can be lowered.
As explained earlier, slightly higher heat rates are also
cost-effective. For example, the base-case design voltage was chosen as
0.8 V per cell. A decrease to 0.787 V per cell increases the current
density and reduces the number of fuel-cell modules by 19%. Further-
more, because more waste heat is available, the temperature-driving
forces in the reformer and heat exchangers E-l, E-2, E-3, and E-4 are
larger, so less heat transfer surface would be required. That would
noticeably reduce investment cost. The resulting increase in heat rate
would increase the fuel cost by only 1.6%.
The combined effect of relaxing the constraints will clearly
lower the optimum cost of electricity, primarily by lowering the invest-
ment cost. However, the exact calculation of that optimum would require
substantial additional effort, requiring analysis of several cases.
Molten carbonate fuel-cell technology is at an early stage of
development, but performance improvements and cost reductions can be
projected for most system components. These expected reductions will
result from current and future R&D programs and system optimization
studies. The impact of these potential improvements was assessed by
assuming the following:
o Improved fuel-cell designs, resulting in a 50% increase in
current density, hence power density, at the design cell
voltage.
o A 15% reduction in the quantity of materials used in stack
construction.
VII-50
-------
o Cost reductions for specific components, including:
Reformer/heat exchanger package 25%
Equipment module 15%
Condenser (E-7) 10%
Power conditioner 20%
Electrical wiring, etc. 15%
o A 15% reduction across-the-board in installation-related costs.
Simultaneous achievement of all cost reduction projections
would lower the installed cost of the System 2 power plant by 27% from
$522/kW to $380/kW. The optimistic projection is listed in Table
VII-23, along with the additional capital requirements for a completely
installed and operating plant.
Operating and maintenance (O&M) costs for operating the power
plant must also be estimated. Firm bases for such estimates are not
available. Periodic stack replacement costs could range from
0.2 to 0.4 cent/kWh, excluding replacement labor costs. This estimate
assumes 40,000 hr operating life at full-rated load (24.0 MW). Fuel
conversion catalyst replacement costs should be low. Equipment
maintenance and replacement costs should also be low, reflecting, say, a
20-year expected life. Thus, total O&M costs could be taken as
0.5 cents/kWh. Finally, local ordinances may require attended operation
of fuel-cell power plants.
The cost of producing electricity from the nominal 26-MW
fuel-cell power plant is shown in Table VII-24. The delivered cost of
SNG is based on the production, pipeline, and distribution costs
presented in previous sections. Capital-related charges are based on
the optimistic power plant costs given in Table VII-23. The O&M cost is
that discussed in the preceding paragraph.
VII-51
-------
Table VII-23
OPTIMISTIC COST PROJECTION FOR NOMINAL 26-MW FUEL-CELL POWER PLANT
Total
Cost,
Item $1000
Fuel-cell trailer 1,628
Reformer/heat exchanger
package 1,512
Equipment module 777
Condenser (E-7) 216
Power conditioner 1,286
Electrical wiring, controls,
and instrumentation 454
Total Manufactured Cost 5,873
Site preparation 409
Freight and insurance 59
Mechanical structures,
foundations, and piping 250
Total Installation Costs 718
Other indirect costs + profit 2,529
Total Power Plant Installed Cost 9,120
Land 20
Interest during construction 460
Working capital 240
Start-up costs 90
Total Capital Investment 9,930
VII-52
-------
Table VII-24
OPERATING COSTS AND REVENUE REQUIREMENTS
FOR A NOMINAL 26-MW FUEL-CELL POWER PLANT
Operating Costs
SNG fuel at $3.98/GJ
($4.20 per million Btu)
Operation and maintenance
Administrative expense
Property taxes and insurance
Depreciation
Total Annual Operating Costs
Return on rate base and income tax
Total Revenue Required
$Mi 11ion/Year
2.22
0.37
0.18
0.23
0.48
3.48
0.87
4.35
Mills/kWh
30.2
5.0
2.4
3.1
6.5
47.2
11.8
59.0
Sources of Revenue
Electric power
4.35
59.0
Figure VII-12 shows the sensitivity of the cost of electricity
to changes in fuel costs and power plant capital costs. Use of the
capital costs estimated in Table VII-22 would increase the cost of
electricity to 69.4 mills/kWh.
L. 26-MW Fuel-Cell Power Plant (Naphtha)
Investment cost estimates for the System 3 power plant were made
based on the detailed evaluation of System 2 component costs. Adjust-
ments were made to reflect differences in size and performance. A
breakdown of the reformer package cost is given in Table VII-25.
VII-53
-------
90
1.0
2.0
SNG COST - dollar per 106 Btu
3.0 4.0 5.0
6.0
7.0
80
70
E
I 60
>
o
cc
o
01
m 50
o
40
OPTIMISTIC CASE
30
20
_L
1.0
2.0 3.0 4.0
SNG COST - dollar per GJ
5.0
6.0
7.0
50
75 100 125
POWER PLANT CAPITAL COST - percent of optimistic case
150
FIGURE VII-12. SENSITIVITY OF THE COST OF ELECTRICITY TO POWER
PLANT CAPITAL COST AND SNG COST
VII-54
-------
Table VII-25
COST BREAKDOWN OF NAPHTHA REFORMER PACKAGE
Cost Item Total Manufactured Cost, $1000
Reformer tubes 135.0
Exchanger tubes
E-l 73.0
E-2 81.7
E-3 51.9
E-4 120.5
Tube manifolds 190.8
Catalyst 19.3
Burner 22.4
Structures 116.2
Total 806.2
The cost of the reformer package is high compared with reformer cost
estimates reported in the literature because it contains heat exchangers
E-l to E-4, which make up 57% of the total cost. The reformer portion
alone would cost about $35/kW for either System 2 or 3.
A summary of component costs for the equipment module is shown in
Table VII-26. The reduction in the size of heat exchanger E-6 accounts
for the lower cost of the equipment module, compared with System 2. The
cost for both systems was reduced by replacing the shell and tube heat
o
exchangers E-5 and E-6 by canal type units, which cost around $161/m
($15/ft2) compared to $592/m2 ($55/ft2) for the shell and tube
configuration.
A breakdown of the estimated plant cost is given in Table VII-27
(see Section VII-K for costing approach). Total installed cost for the
non-optimized base-case System 3 power plant is $13,740,000, equivalent
to $537/kW of net output (25.6 MW). This bottom-line cost is about the
VII-55
-------
Table VII-26
EQUIPMENT MODULE COST BREAKDOWN
Component Total Manufactured Cost, $1000
Exchanger
E-5 14.0
E-6 46.5
Blowers:
B-l 13.3
B-3 47.0
B-4 21.0
\
B-5 23.0
B-6 0.5
Shift reactor 2.0
Hydrodesulfurizer 2.0
ZnO bed 2.5
Pump: P-l 2.8
Knockout 1.7
Module fabrication 53.4
Total 229.7
VII-56
-------
Table VII-27
NOMINAL 26-MW FUEL-CELL POWER PLANT COST ESTIMATE
Item
No. in
System
8
Fuel-cell trailer
Reformer/heat exchanger
package
Condenser (E-7)
Equipment module
Power conditioner and
switchgear 4
Electrical wiring, controls
and instrumentation
Total Manufactured Cost
Site preparation
Freight and insurance
Mechanical structures,
foundations, and piping
Total Installation Costs
Other indirect costs + profit3
Total Power Plant Installed Cost
Unit
Cost,
$1000
284
75
Total
Cost,
$1000
2,296
4
4
4
806
75
230
3,225
300
920
1,710
560
9,011
505
90
308
903
3,830
13,740
Taken as 40% of total manufactured cost + 25% of total installation costs.
VII-57
-------
same as that estimated for the System 2 power plant. The reformer
package cost for the naphtha system is higher. However, this is
counter-balanced by a much lower fuel-cell trailer cost, resulting from
the selection of a lower design voltage that yields higher power density:
o System 2 (SNG): 0.8 V/cell @ 120 mA/cm2 =96.0 mW/cm2
o System 3 (Naphtha): 0.78 V/cell @ 165 mA/cm2 = 128.7
mW/cm2
As before, opportunities exist for major cost reduction in the
System 3 power plant. The impact of future improvements in cell design
and performance and systems concepts was assessed, using the projected
cost reduction factors presented in Section VII-K. Here, it was assumed
that improved cell design would result in an increase in power density
of 40%, rather than 50%. The power density estimated for System 3
already reflects improvements due to the selection of a more favorable
design point.
Simultaneous achievement of all cost reduction projections would
lower the installed cost of the System 3 power plant by 25%, from
$537/kW to $404/kW. The optimistic projection is shown in Table
VII-28.
As with System 2, plant O&M costs are assumed to be about
0.5 C/kWh, based on a target stack life of 40,000 hr and routine
catalyst bed replacement.
The cost of producing electricity from the power plant is shown in
Table VII-29, based on the optimistic capital cost estimate and a load
factor of 35%. The cost of delivered naphtha includes production, pipe-
line, and truck delivery costs.
Figure VII-13 shows the sensitivity of the cost of electricity to
fuel costs and capital costs. If the capital cost estimate given in
Table VII-27 is used, the cost of electricity increases to
69.9 mills/kWh.
VII-58
-------
Table VII-28
OPTIMISTIC COST PROJECTION FOR NOMINAL 26-MW FUEL-CELL POWER PLANT
Item
Fuel-cell trailer
Reformer/heat exchanger package
Equipment module
Condenser (E-7)
Power conditioner
Electrical wiring, controls,
and instrumentation
Total Manufactured Cost
Site preparation
Freight and insurance
Mechanical structures,
foundations, and piping
Total Installation Costs
Other indirect costs + profit
Total Power Plant Installed Cost
Land
Interest during construction
Working capital
Start-up costs
Total Capital Investment
Total
Cost,
$1000
1,394
2,419
782
270
1,368
476
6,709
432
67
262
761
2,874
10,340
20
520
140
100
11,120
VII-59
-------
Table VII-29
OPERATING COSTS AND REVENUE REQUIREMENTS
FOR A NOMINAL 26-MW FUEL-CELL POWER PLANT
Operating Costs $l>000/Year Mills/kWh
Naphtha fuel at $4.01/GJ
($4.23 per million Btu) 2.33 29.7
Operation and maintenance 0.39 5.0
Administrative expense 0.21 2.7
Property taxes and insurance 0.26 3.3
Depreciation 0.55 7.0
Total Annual Operating Costs 3.74 47.7
Return on rate base and income tax 0.98 12.5
Total Revenue Required 4.72 60.1
Sources of Revenue
Electric power 4.72 60.1
VII-60
-------
90
1.0
NAPHTHA COST - dollars per 106 Btu
2.0 3.0 4.0 5.0
6.0
7.0
1 T
80 —
70
60
o
E
U.
O
8
50
40
OPTIMISTIC CASE
BASE CASE
30
20
I
I
1.0
2.0 3.0 4.0 5.0
NAPHTHA COST - dollars per GJ
6.0
7.0
50
75 100 125
POWER PLANT CAPITAL COST - percent of optimistic case
150
FIGURE VII-13. SENSITIVITY OF THE COST OF ELECTRICITY TO POWER
PLANT COST AND NAPHTHA COST
VII-61
-------
M. Electricity Transmission and Distribution
As in the case of gas distribution, the cost of transmitting and
distributing electricity varies widely, and it is extremely difficult to
construct such costs for a particular situation. However, average costs
of transmission and distribution (T&D) can be computed from numerous
statistics. Using these statistics we calculated T&D costs for two
cases. The first represents a typical utility situation in which
electricity is generated in central power stations, transmitted via high
voltage power lines to substations, from which it is distributed to
individual residences. The second case represents the situation in
which electricity is generated in dispersed fuel-cell power plants that
are connected at the substation level, thus eliminating a substantial
part of the transmission cost (interties to the rest of the grid are
still required for reliability, however). Distribution is the same as
in the first case.
The T&D cost is paid by residential customers in two components.
The first is a fixed monthly charge that represents fixed costs to the
utility that are not related to the amount of electricity used —
metering, power poles, billing, and so on. The second is more or less
proportional to the amount of electricity used. These components can be
estimated by examining statistics of the Federal Power Commission
(FPC) and the Edison Electric Institute7 (EEI). According to EEI
statistics the average residential cost of T&D in 1975 was 15.7 mills
per kWh. In the same year, the average residential use of electricity
was 681 kVJh per month. Therefore, the average monthly T&D charge was
$10.67.
The average fixed charges for residential customers may be deter-
mined from data on average monthly residential bills compiled by the
FPC. Using the average of data for January 1, 1975 and applying
least squares analysis to nationwide average electric bills as a func-
tion of the electricity used, the following formula may be derived:
VII-62
-------
C = $3.59 + $0.0286 E (1)
where C is the monthly charge for electricity and E is the amount of
electricity used in kWh. (Analysis of data for utilities in Kansas
City, Omaha, and Des Moines yields a result not significantly different
from the above formula.) Thus, the average monthly fixed charge of
$3.59 accounts for about one-third of residential T&D costs. The
remaining portion that is included in the term proportional to the
amount of electricity used may be calculated as follows:
Since the total average month T&D charge is $10.67, the following
equation must be satisfied:
$10.67 = $3.59 + 681 y (2)
where y is the portion of the electrical rate that represents T&D
charges. Solving this equation yields y = $0.0104/kWh. Thus, the
national average charge for T&D in 1975 can be represented as:
CT&D = $3'59 + $0-0104 E (3)
Using FPC statistics for monthly electrical bills on January 1,
1977 to update Equation 1, we arrive at the following formula:
C = $3.57 + $0.0345 E (4)
Thus, while the fixed portion of residential electricity charges re-
mained essentially unchanged between 1975 and 1977, the variable portion
increased by 21%. To determine the 1977 equivalent of Equation 3, one
must know the proportion of the increase in the variable charge rate due
to generation cost increases and that due to T&D cost increases.
According to EEI, fuel costs alone accounted for 40% of the increase in
the average cost of all electricity in 1975, and 50% in 1976. EEI
figures also show that the average cost of electricity was 2.70
cents/kWh in 1975 and 3.22 cents/kWh in 1977- Using an average
VII-63
-------
percentage increase due to increased fuel costs of 45%, the net increase
due to fuel costs alone was 0.45 (3.22 - 2.70) = 0.23 cents/kWh. Thus,
of the increase in variable charge rate for residential electricity of
(3.45 - 2.86) = 0.59 cents/kWh, 0.23 cents can be attributed to fuel
cost increases. The remaining 0.36 cents must be accounted for by other
generation and T&D cost increases.
The most straightforward way of allocating cost increases is to
scale them according to the ratio of the T&D and generation (minus fuel
costs) components of the variable electricity charge rate in 1975. The
T&D component was previously determined to be 1.04 cents/kWh. The
generation component is then (2.86 - 1.04) = 1.82 cents/kWh. In 1975,
fuel costs accounted for 0.93 cents/kWh of the cost of electricity, on
the average. Therefore, nonfuel generation costs were
(1.82 - 0.93) = 0.89 cents/kWh. Allocating the nonfuel 1975-1977 cost
increase of 0.36 cents/kWh to variable T&D and nonfuel generation in the
ratio of 1.04/0.89 results in a variable T&D cost increase of 0.19
cents/kWh and a nonfuel generation cost increase of 0.17 cents/kWh.
Thus, the cost equation for monthly residential T&D charges for 1977 is:
CT & D = $3'57 + $0.0123 E (5)
The average monthly electricity use per residential customer in 1977 was
729 kWh, so that the average T&D cost was increased 9.5% over 1975:
CT&D = (3'57 + °-0123 x 729)/729 = 1.72 cents/kWh.
Equation 5 represents the average cost of T&D for the typical
utility situation (Case 1) discussed at the beginning of this section.
To calculate the costs appropriate to a distribution of electricity from
dispersed fuel-cell power plants, Equation 5 must be broken into
components of transmission and distribution, and estimates of transmis-
sion cost savings applied to the transmission component.
Q
Bottaro and Baughman have estimated the average national break-
down of residential T&D costs per kWh as follows:
VII-64
-------
Transmission: 32.2%
Distribution: 50.5
General: 17.3.
The "general" category applies to office and overhead expenses that can-
not be allocated specifically to transmission or distribution. Applying
these percentages to the average residential T&D cost of 1.72 cents/kWh
in 1977 results in the following charges:
Transmission: 0.55 cents/kWh
Distribution: 0.87
General: 0.30.
By convention, we have allocated the fixed portion of the monthly charge
rate in Equation 5 to distribution. This part amounts to $3.77/729 =
0.49 cents/kWh on the average. The remaining 0.38 cents/kWh
distribution charge must then be included in the variable portion of
Equation 5, as must the transmission and general charge components.
The savings in transmission charges resulting from employing dis-
persed fuel-cell power plants may be obtained from the work of Wood et
o
al. They estimated that the total capital cost of transmission
installed between 1975 and 1985 would be $166/kW of new generating
capacity, including $26/kW for transmission substations and $40/kW for
subtransmission. They also estimated that for every kW of fuel-cell
capacity that replaces a kW of central station capacity, $29-36/kW of
transmission costs could be saved if the fuel-cell power plants were
connected on the low voltage side of transmission-supplied substations
(appropriate for 26-MW fuel-cell power plants). Using the convention of
g
Bottaro and Baughman, subtransmission lines are included in the
distribution system. Therefore, of a total of $126/kW investment for
transmission, $29-36/kW or 23-29% of transmission capital costs can be
saved by employing dispersed fuel cells.
VII-65
-------
According to Bottaro and Baughman's statistics, 0.51 cent/kWh of
average transmission costs are equipment related. Thus, if about 25% of
these costs can be saved, the resulting savings is 0.13 cent/kWh. Thus,
the T&D cost equation appropriate for dispersed fuel-cell power plants
is Equation 5 with 0.13 cents/kWh subtracted from the variable portion
of the rate, or
$3.57 + $0.0110 E. (6)
The above calculations have used national average data on T&D costs
rather than data for the West North Central states. However,
examination of Bottaro and Baughman's results broken down on a regional
basis indicates that the figures for the West North Central states are
very close to the national average.
N. 100-kW Fuel-Cell Power Plant
Cost estimates were prepared for the 100-kW power plant based on
the equipment specifications discussed in Section IV-E. Costs were
estimated by determining the actual quantities of raw materials used to
fabricate the components. Current costs of these materials in the form
expected to be used were obtained from vendor contacts. Quotes were
based on large quantity purchases. Fabrication costs were determined bj
multiplying the raw material cost by a manufacturing-cost factor, which
was selected according to the production rates involved and the degree
of automation envisioned for the manufacuring facility. The production
rate for cells was assumed to be sufficient to justify the development
and utilization of continuous fabrication processes for the cell compo-
nent parts.
The cost breakdown for fuel-cell stacks is given in Table VII-30.
Total stack cost for the 100-kW power plant is estimated at $20,000.
Catalyst cost is a major factor, based on the initial acquisition cost
of platinum.
VII-66
-------
Table VII-30
FUEL-CELL STACK COSTS
(Production Rate = 1,000 Systems/Year)
Component/Configuration
Catalyst - Platinum
Total loading/cell -
0.001 g/cm2 (0.002 Ib/ft2)
Electrode support layers
Graphite fiber paper -
0.024 g/cm2 (0.05 lb/ft2)
Electrolyte matrix
Silicon carbide fiber -
0.039 g/cm2 (0.08 lb/ft2)
Bipolar plate - carbon/
phenolic resin -
0.44 g/cm2 (0.9 lb/ft2)
Cooling cartridge - carbon plate
with copper tube grid
Stack hardware - end plates,
manifolding, tie rods
Total
Raw
Material
Cost,
$1,000
9.3
2.5
1.07
0.62
0.54
0.71
Mfg.
Cost
Factor
0.05
0.6
0.6
1.5
1.5
1.4
Mfg.
Cost,
$1.000
0.47
1.5
0.64
Total
Cost,
$1,000
9.77
4.0
1.71
14.74
0.93
0.82
0.99
1.55
1.36
1.70
5.35
20.09a
aTotal for 100-kW power plant (4 stacks).
VII-67
-------
The cost of the catalyst is obviously dependent on the base price
of platinum, which fluctuates. The price of $6/g, used for this study,
includes a processing charge of $0.39/g over a base material price of
$5.61/g. At least 80% of this material cost could be reclaimed when the
cell stacks are replaced.
Graphite fiber paper is the major cost component in the electrode
support substrate. According to one vendor contact (Stackpole Carbon
Co.), today's price of $88/kg ($40/lb) inludes a processing charge of
$18 to 22/kg ($8 to 10/lb), for a total market of 1.1 million kg
(0.5 million Ib) per year. A future price of $62 to 66/kg
($28 to 30/lb) could be expected if the market increased to
9 to 11 million kg (4 to 5 million Ib) per year. Other applications,
such as graphite-filled automobile body panels, are expected to help
achieve this forecast market.
Low-cost methods are presently being developed for producing sili-
con carbide fibers for the electrolyte matrix. Ultimately, the cost of
these fibers is expected to approach $13.20/kg ($6/lb), but this is an
optimistic projection. This study assumes a cost of $17.60/kg ($8/lb)
for silicon carbide.
The bipolar plates are produced by a compression molding process
using a mixture of graphite and phenolic powders. This relatively high
cost procedure results in a rather high manufacturing cost, even though
the raw material cost is low — $0.90/kg ($0.41/lb).
The total stack cost is reasonable and not overly optimistic. The
estimated cost is somewhat higher than reported by ERG, primarily be-
cause of the following reasons:
o Selection of a lower cell performance characteristic and
design point, yielding a power density of 727 W/m2
(67.6 W/ft2). This increased the required cell area by 33%.
o Use of a higher current value for platinum catalyst cost.
VII-68
-------
o Incorporation of cooling grids costing $6.60 each, as part of
the circulating coolant system for heat recovery from the
stack.
Clearly, improvements in catalyst utilization and projected cell per-
formance are prime areas for cost reduction.
The final cost estimate for the 100-kW power plant is shown in
Table VTI-31. The estimate is based on large quantity purchases of raw
materials. Major components such as fuel-cell stacks, reformers, and
heat exchangers could be manufactured in separate facilities or pur-
chased. All components are assembled into a single unit at an assembly
plant at an assumed production rate of 1000 units/yr. This facility
would include areas for fabrication and welding of steel structures, and
a small assembly line. System piping and wiring could be fabricated.
The total manufacturing cost is estimated to be $46,610, which is
equivalent to $466/kW output at rated power. The projected FOB selling
price is $65,254 or $652/kW, assuming a mark-up of 40% for other in-
direct expenses and profit.
The nonoptimized base-case design study used conservative estimates
of fuel-cell performance. The impact of future optimization and per-
formance improvement programs on system cost was assessed. The
optimistic projection would be consistent with the later stages of the
assumed 1980-2000 time frame established for this study. Cost reduc-
tions might be needed if systems are to meet the 40,000-hr life goal.
Major improvements can be projected in the area of fuel-cell stack
design, construction, and performance including:
o A reduction in platinum catalyst loading to 0.75 mg/cm2
(0.0015 lb/ft2), together with a doubling of current density
(at 0.65 cell voltage) to 224 mA/cm2 (208 amp/ft2). In
effect, only two stacks would be required for the 100-kW power
plant.
o A 15% reduction in the quantity of materials used in stack
construction.
VII-69
-------
Table VII-31
COST SUMMARY FOR 100-kW POWER PLANT
Unit
Fuel-cell stacks
Reformer
Heat exchangers
E-1A
E-1B
E-2A
E-2B
E-3
E-4
E-5
E-6
E-7
E-8
Blower B-l
Pumps
P-l
P-2
H.T. shift
Cat. G-3A
L.T. shift
Cat. G-66A
Inverter
Instrumentation
Enclosure
Piping, wiring, miscellaneous
Assembly & test
Total Manufacturing Cost
Other indirect costs + profit (40%)
F.O.B. selling price
Freight and insurance
Installation
Total Installed Cost
Cost ($)
20,000
3,800
400
120
150
200
80
1,500
5,700
400
300
600
1,600
240
40Q
300
600
200
1,120
6,000
400
800
500
1,200
46,610
18,644
65,254
650
2.500
68,400
VII-70
-------
o A 15% reduction across-the-board in the manufactured cost of
the reformer, shift converters, heat exchangers and miscel-
laneous cost items (e.g., instrumentation).
o A 20% cost reduction for the DC/AC inverter and associated
power conditioning equipment.
Simultaneous achievement of all cost reduction projections would
lower the FOB price of the 100-kW power plant by 34%, from $652 to
$430/kW. Additional savings could be expected if platinum catalyst
recovery and recycle is carried out. The optimistic projection is
listed in Table VII-32.
Table VII-32
OPTIMISTIC COST PROJECTION OF ADVANCED 100-kW POWER PLANT
Unit Cost ($)
Fuel-cell stacks (2) 8,050
Reformer 3,230
Heat exchangers 8,032
Blower 1,600
Pumps 640
Shift converters 1,887
Inverter 4,800
Miscellaneous3 «/./= =
Total Manufacturing Cost
Other indirect costs + profit (40%)
FOB selling price
Freight and insurance
Installation
Total Installed Cost
Instrumentation, enclosure, piping, wiring, assembly, and testing.
VII-71
-------
This study has focused on costs associated with manufacturing a
single 100-kW power plant product line. If market surveys show that
larger power plants are required, some further cost reduction might
occur, based on capacity factor scale-up relationships, e.g., the 0.6
factor rule. Alternatively, added multiples of 100-kW units might be
installed. This approach would probably be most cost-effective, after
system reliability and redundancy factors are analyzed.
O&M costs for operating the 100 kW power plant also must be esti-
mated, but firm bases for such estimates are not available. Periodic
stack replacement material costs could range from 0.2 to 0.5 c/kWh,
assuming 40,000-hr operations at full rated level (100 kW). Fuel con-
version catalyst replacement costs should be low. Equipment maintenance
and replacement costs should also be low, reflecting, say, a 15-year
expected life. Finally, local ordinances may require attended operation
of total energy power plants.
The cost of producing electricity and heat from the 100-kW power
plant depends strongly on the particular application in which it is
used. Table IV-33 illustrates a particular case in which the load
factor for electricity production is 35%, and 75% of the recovered heat
can be utilized, with an assigned value of $2.84/GJ ($3.00 per million
Btu). The actual costs for the application in System 5 must be deter-
mined from a calculation of annual heat and electricity use. That
calculation is carried out is Chapter VIII. Figure VII-14 shows the
sensitivity of the cost of electricity to load factor and hot water
credit.
0. Hot Water Distribution System
The cost of the system for delivering 82°C (180° F) hot water
from the 100-kW fuel-cell power plant to 20 townhouses represents a sub-
stantial portion of the total energy supply cost for the housing complex.
Table VII-34 shows a breakdown of those costs, which have been estimated
VII-72
-------
Table VII-33
OPERATING COSTS AND REVENUE REQUIREMENTS FOR A 100-kW
FUEL-CELL POWER PLANT WITH HEAT RECOVERY
(35% Load Factor; 75% of Recovered Heat Utilized)
Operating Costs
SNG fuel at $4.16/GJ
($4.39 per million Btu)
Operation and maintenance
Administrative expense
Property taxes and insurance
Depreciation
Total Annual Operating Costs
Return on rate base and income tax
Total Revenue Required
Sources of Revenue
Recovered heat at $2.84/GJ
Electric power at 70.1 mills/kWh
Total
$1,OOP/Year
14.4
1.5
0.9
1.1
2.3
20.2
4.0
24.2
2.7
21.5
24.2
Mills/kWh
46.9
5.0
3.0
3.7
7.5
65.8
13.0
78.8
8.7
70.1
78.8
VII-73
-------
100
1.0
HOT WATER CREDIT - dollars per 106 Btu
2.0 3.0 4.0 5.0
T
6.0
~T
7.0
90 —
I
t- 80
I
t
O 70
u.
O
fe
8
60
LOAD FACTOR
50
40
1.0
2.0 3.0 4.0 5.0
HOT WATER CREDIT - dollars per GJ
6.0
7.0
0.20
0.25
0.30
0.35
LOAD FACTOR
0.40
0.45
0.50
FIGURE VII-14. SENSITIVITY OF THE COST OF ELECTRICITY TO LOAD
FACTOR AND HOT WATER CREDIT
VII-74
-------
Table VII-34
CAPITAL INVESTMENT REQUIRED FOR A SYSTEM THAT DELIVERS
82°C (180°F) HOT WATER TO TWENTY TOWNHOUSES
Investment
Cost Component ($1,000) Percent
Materials
5-cm diameter pipe (46m) 0.25
3-cm diameter pipe (850m) 2.9 5
2-cm diameter pipe (610m) 1.2 2
Pipe fittings 14.5 24
Pipe insulation (1510m) 4.4 7
Pumps (2) 4.5 8
Compression tank 0.42 1
Hot water coils - space heating (20) 4.5 8
Hot water coils - DHW (20) 3.0 5
Total Materials 35.7 60
Labor 8.6 15
Other (permits, trenching, etc.) 2.5 4
Subtotal Direct Costs 46.8 79
Overhead (15% of direct costs) 7.0 12
Profit (10% of direct costs & overhead) 5.4 9
Total Capital Investment 59.2 100
VII-75
-------
using standard construction cost estimation procedures. The size of the
piping, fittings and valves was based on a maximum hot water flow rate
of 1.40 liter/sec (22.1 gal/min) from the fuel-cell heat recovery system.
Included in the cost of the system are the costs of the heating coils
that transfer heat from the hot water stream to the DHW tank and to the
space heating air in each residence. The single most costly expense
category is pipe fittings. This includes the valves that control the
flow of hot water from the fuel cell to the residences and back, and all
elbows, connectors, and so on. Total cost of the heat delivery system,
$59,200, is higher than that of the fuel-cell power plant, and amounts
to nearly $3,000 per residence.
The annual ownership cost of this system is assumed to be borne by
the owners of the townhouses, and is shared equally among the 20 resi-
dences. Financing is assumed to be included in home mortgages of the
individual owners. Financing terms are as follows: 20% down payment,
with the remaining 80% financed over 30 years at 10% interest. The
annual cost of such a loan (principal repayment plus interest) is given
by the following formula:
P i
(1 + i)n - 1
where C is the annual cost, P is the principal, i is the interest rate,
and n is the term of the loan in years. For a principal
P = 0.80 x $2,960 = $2,368, and the loan terms given above, the
homeowner's annual cost is $251. To obtain the average annual cost, the
down payment of $592 is averaged over the 30-year term of the loan, or
$20 per year. The average annual cost, per residence, of the heat
delivery system is therefore $271.
P. Gas Furnace and Air Conditioner
The costs of the gas furnace and air conditioner that supply heat-
ing and cooling to the residences described in Section IV-A are based on
VII-76
-------
published estimates of the costs of the specific models discussed.10
Included are estimates of the cost of duct work for delivering the
heated or cooled air to the various rooms of the residence and venting
for the exhaust gases from the furnace.
The total installed cost of the gas furnace/air conditioning system
is shown in Table VII-35. The equipment costs are based on the whole-
sale price typically paid by a building contractor. The total cost in-
cludes labor for installing the equipment plus the contractor's overhead
and profit.
The annual homeowner's cost of owning this equipment, based on the
considerations discussed in the previous sections, amounts to $195. In
addition, Westinghouse has estimated the annual average maintenance
cost of the equipment to be $99 per year, based on statistics compiled
by gas and electric utilities. Therefore, the total yearly cost of
owning and maintaining this heating and air conditioning system is $294.
Table VII-35
CAPITAL COST FOR A RESIDENTIAL HEATING AND COOLING
SYSTEM — GAS FURNACE AND AIR CONDITIONER
Cost Component Investment, $ Percent
Equipment
Gas furnace (70 MJ/hr) 225 11
Air conditioner (32 MJ/hr) 498 23
Duct work 704 33
Subtotal Equipment 1,427 67
Labor 257 12
Subtotal Labor & Equipment 1,684 79
Overhead (15% labor & equipment) 253 12
Profit (10% labor, equipment & overhead) 194 9
Total Capital Investment 2,131 100
VII-77
-------
Q. Heat Pumps
The costs of the advanced heat pumps described in Sections IV-B and
IV-E were estimated by Westinghouse using a heat pump model that costed
each component individually, then added the cost of all components to
obtain the total. The total installed costs of the two heat pumps
are shown in Table VII-36. As before, the equipment costs are based on
those paid by a building contractor.
Annual average ownership costs for the heating and cooling systems
are $231 for the 26.0-MJ/hr (24,600-Btu/hr) system and $194 for the
19.3-MJ/hr (18,300-Btu/hr) system. Based on heat pump statistics
gathered by electric utilities, Westinghouse estimates the yearly
average maintenance cost to be $120 for either heat pump. Thus, the
total annual maintenance and ownership costs are $351 and $314 for the
26.0-MJ/hr and 19.3-MJ/hr systems, respectively.
Table VII-36
CAPITAL COSTS FOR RESIDENTIAL HEATING
AND COOLING SYSTEMS ~ HEAT PUMPS
26 MJ/hr Heat Pump 19.3 MJ/hr Heat Pump
Cost Component Investment, $ Percent Investment, $ Percent
Equipment
Heat pump
Duct work
Subtotal Equipment
Labor
Subtotal Equipment
& Labor 1996 79 1679 79
Overhead (15% of labor &
equipment) 299 12 252 12
Profit (10% of labor,
equipment & overhead) 230 9 193 9
Total Investment 2525 100 2124 100
VII-78
1138
621
1759
237
45
25
70
9
844
598
1442
237
40
28
68
11
-------
R. References—Chapter VII
1. R. P. Stickles, et al., "Assessment of Fuels for Power Generation
by Electric Utility Fuel Cells," Electric Power Research Institute
Report 318 (October 1975).
2. American Gas Association, "Gas Facts, 1975" (1976).
3. J. M. King, Jr., "Energy Conversion Alternatives Study - United
Technologies Phase II Final Report," NASA CR 134955, FCR-0237
(October 19, 1976).
4. S. Abens, et al., "High Temperature Molten Carbonate Fuel-Cells,"
Fourth Quarter Technical Progress Report E-3-4 (March 1977) and
Fifth Quarter Technical Progress Report E-3-5 (July 1977).
5. P. Wood, "AD/DC Power Conditioning and Control for Advanced
Conversion and Storage Technology," EPRI 390-1-1 (August 1975).
6. Federal Power Commission, "Monthly Electric Utility Bills" (January
1, 1976 and January 1, 1977).
7. Electrical World, various issues.
8. M. L. Baughman and D. J. Bottaro, "Electric Power Transmission and
Distribution Systems Costs and Their Allocation," IEEE
Transactions on Power Apparatus and Systems, p. 782 (May-June 1976).
9. W. Wood, M. P. Bhavaraju, and P. Yatcko, "Economic Assessment of
the Utilization of Fuel Cells in Electric Utility Systems,"
Electric Power Research Institute Report EM-336 (November 1976).
10. H. S. Kirschbaum and S. E. Veyo, "An Investigation of Methods to
Improve Heat Pump Performance and Reliability in a Northern
Climate," Electric Power Research Institute Report EM-319 (January
1977).
VII-79
-------
VIII. THERMAL AND ELECTRICAL LOAD CALCULATIONS
To complete the analysis of the five systems, the energy use char-
acteristics of the residences described in Chapter IV must be known.
Those residences are the ultimate users of the energy supplied by the
other system components. The determination of their annual energy con-
sumption will allow the costs, energy use, and environmental impacts of
the systems to be appropriately scaled for comparison.
As discussed in Chapter III, our study focuses on the heating and
cooling components of residential energy use. That emphasis is reason-
able, because approximately 60% of residential energy use in the United
States is for heating and cooling. Also, because the energy demand for
heating and cooling varies widely with the time of day, those portions
of the residential energy load tend to contribute substantially to the
utility's "intermediate load" electricity demand, which is the type of
load that the electricity generating components of the systems are de-
signed to meet.
In addition to the heating and cooling demand, the demand created
by other loads in the residences — lights and appliances — must be
known for two reasons. First, the cost per kWh of electricity to the
customer will depend on the total monthly electricity use. Second, in
System 5, the amount of heat supplied by the fuel-cell power plant, and
thus the amount of heat required of the heat pumps, will depend on the
total electrical load of the residences.
The electrical loads we used for lights and appliances are
statistical averages. The heating and cooling loads, however, and their
resultant electricity demand, have been determined on the basis of the
following information: (1) the daily temperature variations in the
VIII-1
-------
Omaha-Des Moines-Kansas City region during a typical year; (2) the
thermal response of the residences to those variations; and (3) the
performance of the heating and cooling equipment as a function of
temperature and the thermal demands of the residence. In the following
sections, the use of those pieces of information to calculate final
residential energy demand is described.
A. Light and Appliance Loads
The electrical loads generated by lights and appliances in a resi-
dence are a function of the number and types of appliances, the number
of occupants, their living pattern, and so on. Although those factors
vary substantially from one household to the next, the quantities of
interest to utilities, as well as to this study, are the average loads
determined by the characteristics of multiple households. Such statis-
tical information is readily available.
The average light and appliance loads will be characterized for the
three types of residences described in Chapter IV. The first, which
will be called Residence 1, is supplied by System 1. It uses both gas
and electricity. In addition to a gas furnace, it is assumed to have a
gas range and water heater. All other appliances are electric, and in-
clude an air conditioner, clothes washer and dryer, refrigerator/freezer,
television, dishwasher, and other small appliances such as a toaster and
food mixer.
The second type of residence, called Residence 2, is supplied by
Systems 2, 3, and 4. Residence 2 is all-electric, and in addition to
having a 26.0-MJ/hr (24,600-Btu/hr) heat pump for heating and cooling,
it employs an electric water heater and range. The use of lights and
other appliances is the same as for Residence 1.
Residence 3, which is supplied by System 5, is similar to Residence 2
except that it employs a smaller heat pump (19.3 MJ/hr or 18,300 Btu/hr)
VIII-2
-------
and its domestic hot water (DHW) is supplied entirely by heat recovered
from the fuel-cell power plant; therefore, no electric water heating is
required.
The components of electricity consumption, by lights and appliances
as well as monthly totals, for the three types of residences are shown
17
in Table VIII-1, as determined from national statistics. ' Those
figures do not include air conditioner or heat pump loads — they will
be determined in following sections.
To determine the ability of the System 5 fuel-cell power plant to
respond to the electrical and thermal loads of the townhouses, one must
know the variations in the light and appliance loads of Residence 3 with
the time of day. Those load variations have been previously esti-
3
mated, and are shown in Figure VIII-1. The loads shown are the
average hourly loads per residence. The actual load profile for an
individual residence would look considerably different, having many
abrupt changes in load as various appliances were turned on and off.
The load profile in Figure VIII-1 is seen to be at a minimum of
0.27 kW in the late night and early morning hours when only the refrig-
erator and perhaps one or two small lights are using power. Peak
demands of 2.83 and 2.46 kW occur during the hours of 9-10 a.m.
Table VIII-1
MONTHLY LIGHT AND APPLIANCE ELECTRICAL
LOADS FOR THREE TYPES OF RESIDENCES
Electricity Use (kWh/Month)
Residence 1 Residence 2 Residence 3
Source
Lights 145 145 145
Water heater — 380
Range — 100 100
Other appliances 490 490 490
Total 635 1,115 735
VIII-3
-------
o
o
3
LIGHTS AND APPLIANCES
10
I
6§
m
Q
•i
12M 1 23456789 10 11 12N1 234 567 89 10 11 12M
AM PM
TIME OF DAY
CD
<
FIGURE VIII-1. VARIATION IN HOURLY AVERAGE LIGHT AND APPLIANCE LOADS AND DHW DEMAND WITH
TIME OF DAY - RESIDENCE 3
-------
and 7-8 p.m., respectively, when the bulk of domestic activities are
taking place.
Also shown in Figure VIII-1 is the DHW demand profile, which will
also be required in the supply/demand calculations for System 5.
B. Daily Temperature Variations
Data on temperature variations for most cities can be obtained from
the Environmental Data Service of the National Oceanic and Atmospheric
Administration. Temperatures are reported every 3 hours, every day of
the year. In addition, daily and monthly averages are reported, as well
as daily extremes, along with various statistical temperature data. To
determine typical seasonal heating and cooling requirements, one must
examine temperatures averaged over many years. Table VTII-2 shows the
normal monthly average temperatures for the period 1941-1970 for Des
Moines, Omaha, and Kansas City.
Table VIII-2
NORMAL MONTHLY AVERAGE TEMPERATURES, °C (°F)
Month
January
February
March
April
May
June
July
August
September
October
November
December
Kansas City
-2
0
5
12
18
23
26
25
20
14
6
1
.3
.6
.1
.8
.3
.3
.0
.2
.4
.8
.4
.7
(27
(33
(41
(55
(65
(73
(78
(77
(68
(58
(43
(32
.8)
.1)
.2)
.0)
.0)
.9)
.8)
.4)
.8)
.6)
.6)
.3)
_e
-2
2
11
17
22
25
24
19
13
4
-2
Omaha
.2
.2
.8
.3
.2
.3
.1
.2
.1
.3
.4
.2
(22.
(28.
(37.
(52.
(63.
(72.
(77.
(75.
(66.
(55.
(40.
(28.
6)
0)
1)
3)
0)
2)
2)
6)
3)
9)
0)
0)
Des Moines
-7
-4
1
9
16
21
23
22
17
12
3
-3
.0
.3
.1
.7
.1
.4
.9
.9
.9
.4
.2
.9
(19.4)
(24.2)
(33.9)
(49.5)
(60.9)
(70.5)
(75.1)
(73.3)
(64.3)
(54.3)
(37.8)
(25.0)
VIII-5
-------
Temperatures for Omaha are a reasonable average of temperatures for
the three cities. Therefore, in all further calculations, only
temperature data for Omaha was used. Monthly temper- ature data for
Omaha for the 10 years including 1966, 1968, and 1970-77 were analyzed
to find particular months that were most representative of normal
monthly conditions (statistics for 1967 and 1969 were not
available). Both monthly average temperatures and monthly heating (or
cooling) degree-days were compared to obtain the best match. Occasion-
ally, when two months were very similar, other data such as monthly min-
imum and maximum temperatures were compared. The results of the match-
ing procedure are shown in Table VIII-3, in which the normal monthly
temperatures and heating (or cooling) degree-days for Omaha are compared
with the average temperatures and degree-days for the actual months
chosen as the best match. Those months are indicated by the year in
which they occur.
Table VIII-3
COMPARISON OF NORMAL MONTHLY CONDITIONS WITH
ACTUAL MONTHLY CONDITIONS OF MONTHS CHOSEN AS "BEST MATCH"
Actual Conditions
Heating
(Cooling)
Average Degree-
Year Temperature (°C) Days (°C)
Normal Conditions
Heating
(Cooling)
Average Degree-
Month Temperature (°C) Days (°C)
January
February
March
April
May
June
July
August
September
October
November
December
-5.2
-2.2
2.8
11.3
17.2
22.3
25.1
24.2
19.1
13.3
4.4
-2.2
730
576
481
217
(48)
(131)
(210)
(186)
(61)
167
417
637
1975
1968
1971
1974
1972
1975
1970
1970
1977
1974
1966
1966
-5.3
-2.6
2.9
11.4
16.9
22.5
25.2
24.6
19.7
12.9
4.5
-2.1
728
603
474
211
(41)
(134)
(214)
(210)
(58)
167
412
629
VIII-6
-------
Table VIII-3 shows that the conditions of the actual months chosen
very closely match the normal temperature condition. Those months are
assumed to constitute a "typical" year for Omaha, and the daily temper-
ature variations were used to calculate heating and cooling loads.
C. Thermal Response of the Residences
Daily temperature data can be used to calculate heating and cooling
loads if the thermal characteristics of the residences are known. A
method for calculating those loads has been developed by Westinghouse
(its report may be referred to for details of the derivation of the
method.) Basically, the thermal response of the residence is calculated
by using an electric circuit analog to derive the appropriate response
functions. The difference between external and internal temperatures is
analogous to voltage, heat flow is analogous to current, and thermal
conductivity and heat capacity are analogous to electrical conductivity
(the reciprocal of resistance) and capacitance, respectively.
The result of that analysis for heating loads is given in Equations
1-3, as follows:
QL<*) " Qavg + AQL(t) (1)
QL(t) = GX ATa(t) + G1(a2 - «i) /Q ATa(t')dt' + C (2)
C = - G! /2* ATa(t)dt - d ( a2 ~ Oil) /2ndt /O ATa(t')dt' (3)
24 ° 24
-°2
48
In Equation 1, QT(t) is the heating load in kJ/hr (or Btu/hr) as a
LI
function of the time of day, t (in hours); Qavg is the daily average
heating load calculated using the average daily temperature and the heat
VIII-7
-------
loss equation discussed in Chapter IV, and AQ (t) is the variable
portion of the heating load to be calculated by Equations 2 and 3. In
Equation 2, G., a., and Ot2 are thermal parameters of the resi-
dence, AT is the difference between the temperature at t and the
Q.
daily average temperature, and C is a constant of integration to be
calculated by Equation 3. In Equation 3, C^ is another thermal
parameter of the residence, T is the daily average temperature of
the day for which the heating load is being calculated, and T .is
the average temperature of the previous day.
It can be seen from Equations 1-3 that calculating the daily aver-
age heating load by integrating over the 24-hour period will result in
exactly Q , modified slightly by the addition of the third term in
Equation 3. That term represents the long-term thermal storage capacity
of the residence. It will also average to zero, however, if the calcu-
lation is carried out over a successive period of days — say, for a
month. Thus, for calculating monthly and seasonal heating loads, it is
clear that the use of Q is sufficient to provide the needed infor-
mation to an acceptable level of accuracy. Recall that for Residences 1
and 2, Q& is given by:
Qavg = 39,900 - 360 Tflvg kJ/hr (4)
or
(Qavg = 37'800 ~ 61° Tavg
where T a is the average daily temperature in C or F. For
Residence 3, the heating load is:
Qavg = 23,800 - 230 T kJ/hr (5)
or
(Qavg = 22'600 - 39° Tavg Btu/hr>-
Because of the complicated relationship between heating load, electrical
load, and electrical and heating supply for the Residence 3 case, one
must consider whether daily average calculations can effectively repre-
VIII-8
-------
sent the averaged effect of hourly load variations. This question is
discussed in the following section.
The calculation of cooling loads is considerably more complicated
than the calculation of heating loads; not only does temperature vary
throughout the day, but so does solar heat input (insolation) and latent
heat infiltration (the latent heat content of humid outside air entering
the house). However, several simplifying assumptions can make the task
easier. First, most of the cooling load is registered during summer
afternoons. Therefore, figures for average summer afternoon insolation
and relative humidity provide reasonable estimates of solar heating and
latent heat infiltration.
Second, the daily average cooling load can be estimated in a way
similar to that for the heating load. Instead of the average daily tem-
perature, however, the average temperature is used for the hours during
which cooling is required. The parameters that contribute to the cool-
ing loads for Residences 1, 2, and 3 are shown in Table VIII-4. The
heat input from people and appliances is similar to that used to calcu-
late heating loads. The heat input from latent heat infiltration was
calculated using a technique given in the ASHRAE handbook, and is
based on an average moisture content of 0.0131 kg HO per kg of air
for exterior air (average summer afternoon conditions for Omaha), 0.0117
kg Ho per kg of air for interior air (corresponding to 26 C wet
bulb interior temperatures). The figures for sensible heat input (via
heat transfer through the walls and infiltration of warm air) and solar
heat input were derived in the Westinghouse report .
Using the parameters listed in Table VIII-4 to calculate the
cooling load as a function of exterior temperature results in the
following equation for Residences 1 and 2:
or
Qavg = 380 Tavg- 43,800 kJ/hr (6)
(Qavg « 650 Tavg - 41,600 Btu/hr)
with Tavg expressed in °C or °F, as appropriate.
VIII-9
-------
To calculate the design cooling loads discussed in Chapter IV, the
moisture content corresponding to the 97.5% temperatures for Omaha
(34°C or 93°F dry bulb and 26°C or 78°F wet bulb) was used —
0.0175 kg HO per kg of air. Use of that quantity yields latent heat
infiltration of 5,700 kJ/hr (5,400 Btu/hr) for Residences 1 and 2 and
2,850 kJ/hr (2,700 Btu/hr) for Residence 3.
The seasonal heating and cooling loads may now be calculated for
Residences 1, 2, and 3 by using the equations developed in this section
and the temperature data for a "typical" year for Omaha. The heating
season is assumed to extend from October 1 through April 30, and the
cooling season from May 1 through September 30. The seasons overlap
somewhat (i.e., some heating is required in May and September, and some
cooling is required in April and October), but it is small and may be
ignored.
Calculation of heating loads is straightforward. The daily average
temperature is used to calculate the daily average heat load for each
day in which the average temperature is below 17°C (62°F) for Resi-
dences 1 and 2, and below 14°C (58°F) for Residence 3. Those are
the temperatures at which the internal heat sources (people, lights, and
appliances) are sufficient to maintain the interior temperature at 21°C
(70 F); above those temperatures the heating load is zero. The daily
average heating loads are then multiplied by 24 and summed over the num-
ber of days in the month to obtain the total monthly heating loads.
Table VIII-4
COMPONENTS OF THE COOLING LOAD — OMAHA SUMMER AFTERNOON CONDITIONS
Component Residences 1 and 2 Residence 3
Heat input from
people, lights,
& appliances 4,960 kJ/hr (4,700 Btu/hr) 4,960 kJ/hr (4,700 Btu/hr)
Solar heat input 3,380 kJ/hr (3,200 Btu/hr) 840 kJ/hr (800 Btu/hr)
Latent heat
infiltration 1,390 kJ/hr (1,310 Btu/hr) 700 kJ/hr (660 Btu/hr)
Sensible heat
input 380 kJ/hr-°C 240 kJ/hr-°C
(200 Btu/hr-c-F) (125 Btu/hr-°F)
VIII-10
-------
The calculation of cooling loads is slightly more complicated.
Only those portions of the day during which the external temperature
exceeds 26 C (78 F) are used in the calculation. The recorded tri-
hourly temperatures are examined to determine the number of hours
during which the temperature exceeded 26°C. The average temperature
during those hours is used to calculate the average cooling load using
Equations 6 and 7. The cooling load is then multiplied by the number of
hours during which the temperatures exceeded 26°C to obtain the total
daily cooling load. The daily cooling loads are then summed to obtain
the cooling loads for each month.
The calculated results for heating and cooling loads are shown in
Table VIII-5. The heating load for either type of residence far exceeds
the cooling load. Thus, the use of heat pumps that are optimized for
heating performance (as described in Chapter IV) is fully justified.
Month
October
November
December
January
February
March
April
Total
May
June
July
August
September
Total
Table VIII-5
SUMMARY OF HEATING AND COOLING LOADS BY MONTH
Residences 1 and 2
Residence 3
Heating Load, GJ
3.52 (3.34)
10.1 (9.62)
16.1 (15.3)
19.0 (18.0)
15.5 (14.7)
11.8 (11.2)
4.80 (4.55)
(Million Btu)
1.32 (1.25)
5.22 (4.95)
9.00 (8.53)
12.6 (11.9)
8.67 (8.22)
6.28 (5.95)
2.00 (1.90)
80.9
(76.7)
45.0 (42.7)
Cooling Load, GJ (Million Btu)
1.03
2.91
5.17
4.48
0.73
(0.98)
(2.76)
(4.90)
(4.25)
(0.69)
0.69 (0.65)
1.92 (1.82)
3.40 (3.22)
2.95 (2.80)
0.49 (0.46)
14.3 (13.6)
9.44 (8.95)
VIII-11
-------
D. Energy Consumption for Heating and Cooling
The final step in the sequence outlined at the beginning of the
chapter is to calculate the response of the heating and cooling equip-
ment to the heating and cooling loads derived in previous sections, and
ultimately to calculate the energy consumed by that equipment. For
Residences 1 and 2, the calculations are relatively straightforward.
The energy consumed by the gas furnace is simply equal to the heating
load divided by the thermal efficiency of the furnace, which is 0.60
(see Chapter V). For the heat pump operating in the heating mode, the
heating load is compared to the heat output of the heat pump at a given
temperature. If the heating capacity of the heat pump exceeds the
heating load, the heat pump will be operating only part of the time.
The fraction of the time the heat pump is on is called the duty factor.
The average electricity demand by the heat pump (in kW) is equal to its
electricity demand at the temperature in question multiplied by the duty
factor.
When the heating load exceeds the capacity of the heat pump, sup-
plemental electric resistance heating must be used to make up the dif-
ference. The heat pump operates continuously and consumes power at its
rated capacity at the temperature in question. The resistance heaters
will consume power at an average rate that will equal the difference
between the heating load and the heat supply of the heat pump. The
total power demand is the power demand of the heat pump plus the average
power demand of the resistance heaters.
Calculation of power demand for the heat pump operating in the
cooling mode and the air conditioner is similar, except that when the
cooling load exceeds the cooling capacity, the duty factor is 1.0 and
the air conditioner or heat pump power demand is equal to its rated
demand at the temperature in question. That is a consequence of there
being no "supplemental cooling capacity." In practice, that means that
at temperatures above the balance point of the cooling device, the inte-
rior temperature of the residence cannot be maintained at the "set" tem-
perature, but will actually be somewhat higher.
VIII-12
-------
In principle, the calculations described above should be carried
out for every hour of the heating and cooling seasons to determine the
total electricity consumption. However, because the heat pump and air
conditioner are reasonably close to being linear in their electricity
demand as a function of temperature, the use of daily average tempera-
tures closely approximates actual energy use.
The heat pump and air conditioner average power demands were used
in conjunction with the daily average temperature data for Omaha's
"typical" year to compute the monthly electricity use for each type of
residence.
Table VII-6 shows the monthly electricity use for heating and
cooling and the total seasonal electricity use, including light and
appliance use of 635 kWh/month for Residence 1, and 1,115 kWh/month for
Residence 2.
Table VIII-6
ELECTRICITY CONSUMPTION FOR HEATING AND COOLING (kWh)
Residence 1
Month
October
November
December
January
February
March
April
Seasonal Total
Month
May
June
July
August
September
Seasonal Total
Annual Total
Heating
27
79
125
148
121
92
37
629
Cooling
137
384
690
594
96
1,901
2,530
L&Aa
635
635
635
635
635
635
635
4,445
L&A
635
635
635
635
635
3,175
7,620
Total
662
714
760
783
756
111
672
5,074
Total
772
1,019
1,325
1,229
731
5,076
10,150
Residence 2
Heating
292
1,025
2,293
2,968
2,108
1,307
413
10,406
Cooling
107
307
552
471
73
1,510
11,916
L&A
1,115
1,115
1,115
1,115
1,115
1,115
1.115
7,805
L&A
1,115
1,115
1,115
1,115
1,115
5,575
13,380
Total
1,407
2,140
3,408
4,083
3,223
2,422
1,528
18,211
Total
1,222
1,422
1,667
1,586
1,188
7,085
25,296
aL&A = light and appliances.
VIII-13
-------
For Residence 1, the electricity used for heating is only that
consumed by the gas furnace's electric fan motor. The total con-
sumption of natural gas (or SNG) by the furnace is 135 GJ (128 million
Btu) over the entire heating season.
The calculation of heating energy use is a more difficult task for
Residence 3 than for Residences 1 and 2 because of the strong functional
relationship between temperature, heating load, and electrical load. To
explore that relationship, we calculated the variations in those param-
eters over a 24-hr day. We considered two major independent variables
— the exterior temperature, and the light and appliance electrical
load. The variations in temperature are taken from the temperature data
for the Omaha "typical year", and the light and appliance loads are
shown in Figure VIII-1. The DHW demand shown in that figure was an
important input to our calculation.
To calculate the heating load that must be met by a combination of
the heat pump and fuel-cell recovered heat, Equations 1, 2, and 3 were
used with the appropriate thermal constants for Residence 3. We inte-
grated those equations numerically, so that heat loads could be calcu-
lated over a 24-hr period.
In addition to the temperature, heat load, DHW load, and light and
appliance electrical load, the inputs required for the program are the
functions that relate heat pump input and output to temperature, fuel-
cell heat recovered to electrical load, and heat transfer efficiency to
heat pump output and fuel-cell heat recovered. The first three are ob-
tained from data presented in Chapters IV and V, and the fourth is ob-
tained from the manufacturer's literature on the Trane WC-18 heating
coil (see Section V-P). The four functions were represented by alge-
braic expressions that were fitted to numerical data.
With all the above inputs formulated, we calculated hourly average
electrical loads, heat pump duty factors, and the amount of recovered
fuel-cell heat delivered to the space heating system. The operation of
the program is described in Appendix C.
VIII-14
-------
The amount of recovered fuel-cell heat available when only lights
and appliances are operating are illustrated along with the DHW demand
in Figure VIII-2. The figure shows that there will always be sufficient
recovered fuel-cell heat to supply all DHW needs for the residences in
System 5, even when the only electrical load is the basic light and
appliance load. Additional heat recovered from the fuel-cell power
plant beyond that required to meet DHW demand will be used to satisfy
the space heating load. When there is no space heating load, the extra
heat is discharged to the atmosphere.
As an example of the performance of the 100-kW fuel-cell power
plant and associated equipment in meeting the electrical and thermal
loads of 20 townhouse residences, as well as of the computational pro-
gram discussed previously, the coldest day of the "typical" year for
Omaha was chosen. The highest electrical demand of the heating season
occurs on that day, so that the capability of the 100-kW fuel-cell power
plant to meet that demand for 20 residences must be determined. The
actual temperature data were for January 12, 1975. On that day the tem-
perature ranged from -23°C (-9°F) to -13°C (8°F) with an average
temperature of -17 C (1.5 F). Figure VIII-3 shows the variation in
temperature with time of day along with the hourly average heating loads
for Residence 3 calculated using Equations 1, 2, and 3. The heating
load closely follows the temperature, with the peak load of 24.8 MJ/hr
(23,490 Btu/hr) occurring at 6-7 a.m. That heating load must be met by
a combination of recovered fuel-cell heat, heat pump output, and elec-
tric resistance heat.
The program that calculates the performance of the energy supply
system for the townhouses was run using the temperature data and heating
loads shown in Figure VIII-3. The results are illustrated in Figures
VIII-3 and VIII-4. Figure VIII-3 shows the portion of the heating load
per residence that is_met by heat recovered from the fuel-cell power
plant. Over the 24-hr period, that amount is 234 MJ (221,600 Btu), or
42% of the total heating load of 558 MJ (529,100 Btu); in addition, the
fuel-cell power plant supplies 44.9 MJ (42,500 Btu) of DHW demand.
VIII-15
-------
13
12
10
HI
Q
cr
o
a.
a.
en
UJ
I
i i i i i r
n i i r
I
12M 1
I I
i r
I I
J L
FUEL CELL HEAT AVAILABLE
12,000
10,000
8000
6000
.n
D
Q
<
III
Q
cr
o
_
D-
<
LU
4000
2000
567
A.M.
10 11 12N 1 2
TIME OF DAY
6
P.M.
8 9 10 11 12M
FIGURE VIII-2. DHW DEMAND RELATIVE TO FUEL CELL HEAT AVAILABLE FROM OPERATION OF LIGHTS AND APPLIANCES
-------
u
V15
UJ
cc
I-
LU
a.
UJ
-25
10
-5
-10
UJ
UJ
D.
w
<
M
H
H
25
I 20
IT 15
O
D
<
jj? 10
O
UJ
I
12M 1
HEAT SUPPLY
25,000
20,000
15,000
3
m
10,000
5,000
6
AM
10 11 12N 1
TIME OF DAY
6
PM
10 11 12M
FIGURE VIII-3. HOURLY AVERAGE TEMPERATURE, HEATING LOAD, AND SUPPLY OF RECOVERED FUEL CELL HEAT
FOR RESIDENCE 3 ON THE COLDEST DAY OF THE YEAR
-------
I I I I I I I i r
i i i
i i i
I
H-'
OO
I
0
3
_i
o
(E
O
21-
TOTAL ELECTRICAL LOAD
ELECTRIC RESISTANCE
HEATERS
LIGHTS AND APPLIANCES
12M 1 23456789 10 11 12N 1 23456789 10 11 12N
AM PM
TIME OF DAY
era
(D
<
FIGURE VIII-4. HOURLY AVERAGE ELECTRICAL LOADS FOR RESIDENCE 3 ON THE
COLDEST DAY OF THE YEAR
-------
Figure VIII-4 shows the hourly average electrical load per resi-
dence, categorized according to lights and appliances, heat pump, and
electrical resistance heat. The figure clearly shows that the average
electrical load does not exceed the 5 kW per residence capacity of the
fuel-cell power plant. However, peak concident electrical demand could
exceed the 5 kW capacity during high demand periods, because the loads
in Figure VIII-4 are hourly averages rather than actual instantaneous
loads.
To determine whether the calculation discussed above would be
required for each day of the heating season to arrive at accurate
monthly and seasonal electricity consumption for System 5, the daily
averages of the heating and electrical loads presented above were com-
pared to those derived from daily average temperature, heating load,
light and appliance load, and DHW demand. Average light and appliance
load and DHW demand of 1.01 kW and 1,870 kJ/hr (1,770 Btu/hr), respec-
tively, along with the average temperature of -17°C (1.5°F) and
space heating load of 23.2 MJ/hr (22,000 Btu/hr), were used to calculate
the following daily average parameters:
o Average electrical load: 3.20 kW
o Recovered fuel-cell heat delivered to the space heating system:
9,810 kJ/hr (9,300 Btu/hr).
The corresponding parameters derived by averaging the hourly values
calculated previously are 3.04 kW and 9,740 kJ/hr (9,230 Btu/hr). These
results, as well as others obtained from daily temperature data for less
severe temperature conditions, indicate that the use of daily average
values in the calculation of average electrical loads and recovered
fuel-cell heat supply yield acceptable agreement (generally within 5%)
with values calculated from 24-hour data. Therefore, the daily average
method was used to calculate monthly and seasonal electricity consump-
tion and fuel-cell heat utilization for the townhouse residences.
Figure VIII-5 shows the heating system electrical load (heat pump
plus electric resistance heaters) and recovered fuel-cell heat delivered
VIII-19
-------
5.0
-20
-10
10
TEMPERATURE - °F
20 30
40
50
60
25
I
N>
O
4.0
SPACE HEATING
ELECTRICAL LOAD
3.0
<
O
O
111
2.0
1.0
DELIVERED
FUEL CELL
HEAT
-30
-20
-10 0
TEMPERATURE - °C
10
20
.*
15 I
t-
Q
uu
DC
10
LU
Q
20
FIGURE VIII-5. VARIATIONS IN AVERAGE SPACE HEATING ELECTRICAL LOAD AND
DELIVERED FUEL CELL HEAT WITH EXTERNAL TEMPERATURE
-------
to the heating system as a function of external temperature. They are
based on an average light and appliance load of 1.01 kW and a DHW load
of 1,870 kJ/hr (1,770 Btu/hr). The electrical load curve shows a
leveling in slope at about 9°C (48°F). That is the point at which
recovered fuel-cell heat is sufficient to supply the heating load of the
residences. Above that point the only electricity requirement is the
0.21 kW consumed by the heat pump fan motor, which must be on whenever
space heat is required. Both curves are remarkably linear considering
the complicated functional relationship between the various system
parameters.
To calculate cooling season energy requirements, the same method
was employed as for Residences 1 and 2, because recovered fuel-cell heat
does not enter into the calculation. To check the peak cooling demand
against the capacity of the 100-kW power plant, temperatures for the
hottest day of the "typical" year (represented by July 1, 1970) were
used to calculate hourly electrical loads. The results of that calcula-
tion are shown in Figure VIII-6. Again, the peak of the average hourly
demand of lights and appliances plus heat pump operating in the cooling
mode does not exceed the 5 kW per residence capacity of the fuel-cell
power plant.
Using the methods discussed above, the monthly and seasonal heat-
ing, cooling, and total electrical consumption for Residence 3 were cal-
culated, along with the amount of recovered fuel-cell heat utilized to
meet the heating load. The results of those calculations are shown in
Table VIII-7. Total use of recovered fuel-cell heat for space heating
is 22.7 GJ (21.5 million Btu) for the heating season. In Section VIII-B
the seasonal heating load for Residence 3 was given as 45.0 GJ
(42.7 million Btu). Therefore, recovered fuel-cell heat supplies 50% of
the total space heating requirement in addition to the total yearly DHW
requirement of 16.3 GJ (15.5 million Btu).
VIII-21
-------
40
O
I
LU
DC
35
O.
30
I i i I
105
100
i
<
95
90
85
80
75
tc.
D
5
UJ
I
to
N5
I
33
O
3
£2
O
III
12M 12 3456789 10 11 12N 1
AM
234 56 789 10 11 12M
PM
FIGURE VIII-6. HOURLY AVERAGE TEMPERATURE AND ELECTRICAL LOAD FOR RESIDENCE 3
ON THE HOTTEST DAY OF THE YEAR
-------
The annual electricity consumption for Residence 3 is 12,790 kWh.
The annual load factor of 0.292 for a 100-kW power plant supplying 20
townhouses is obtained by dividing 20 times 12,790 kWh by the electrical
capacity of the fuel-cell power plant (100 kW x 8,760 hr = 876,000 kWh).
This is somewhat lower than the 0.35 load factor assumed for other types
of power plants in this study. However, it is a reasonable value for a
power plant that is expected to meet all the electrical requirements of
a limited number of residences, ranging from minimum to peak loads.
Table VIII - 7
RESIDENCE 3 ELECTRICITY CONSUMPTION AND FUEL-CELL HEAT UTILIZATION
Utilization of
Recovered Fuel-Cell Heat (GJ)
Electricity Consumption (kWh)
Month
October
November
December
January
February
March
April
Seasonal
Total
DHWa
1.36
1.36
1.36
1.36
1.36
1.36
1.36
9.52
Space Heating Total
1.10
3.12
4.43
4.98
4.06
3.46
1.51
22.7
2.46
4.48
5.79
6.34
5.42
4.82
2.87
32.2
Heating
98
342
598
756
620
415
118
2,947
L&AD
735
735
735
735
735
735
735
5,145
Total
833
1,077
1,333
1,491
1,355
1,150
853
8,092
May
June
July
August
September
Seasonal
Total
Annual
Total
1.36
1.36
1.36
1.36
1.36
6.80
16.30
.36
.36
22.7
1.36
1.36
1.36
6.80
39.0
69
201
375
324
51
735
735
735
735
735
804
936
1,110
1,059
786
1,020 3,675 4,695
3,967 8,820 12,787
aDHW = domestic hot water.
L&A = light and appliances.
VIII-23
-------
E. References—Chapter VIII
1. Gordian Associates, Inc., "Evaluation of the Air-to-Air Heat Pump
for Residential Space Conditioning," Federal Energy Administration
Report FEA/D-76/340 (April 1976).
2. Stanford Research Institute, "Patterns of Energy Consumption in the
United States," Office of Science and Technology (January 1972).
3. Hittman Associates, Inc., "Residential Energy Consumption - Single
Family Housing," Department of Housing and Urban Development Report
HUD-HAI-2 (September 1975).
4. U.S. Department of Commerce, National Oceanic and Atmospheric
Administration, Environmental Data Service, "Local Climatological
Data — Omaha, Nebraska."
5. H. S. Kirschbaum and S. E. Veyo, "An Investigation of Methods to
Improve Heat Pump Performance in a Northern Climate," Electric Power
Research Institute Report EM-319 (January 1977).
6. American Society of Heating, Refrigerating, and Air Conditioning
Engineers, Handbook of Fundamentals (New York 1972).
VIII-24
-------
IX. COMPARATIVE ANALYSIS
Because the cost, efficiency, and environmental parameters for the
various system components have been determined, the overall performance
of each of the five systems can now be analyzed and compared. The sub-
sequent analysis in this chapter compares the relative merits of the
systems rather than measuring them against some absolute standard of
performance. However, to establish a benchmark against which the other
systems can be compared, System 1 is considered to be closest to the
current energy supply situation; other systems are assessed as having
advantages or disadvantages compared to System 1.
All costs, energy consumption, and environmental impacts are ex-
pressed in terms of the total heating and cooling supplied to the resi-
dences. The unit of measure is 1 GJ (0.948 million Btu) of heating and
cooling. A summary of the overall advantages and disadvantages of the
systems is presented in Chapter X.
A. Energy Efficiency
The overall energy efficiencies of each of the five systems are
determined by the thermal efficiency factors presented in Chapter V.
For several system components, however, efficiencies were calculated as
a function of percent of rated load (fuel-cell power plants, combined
cycle power plants) or as a function of exterior temperature (heat
pumps, air conditioners). Therefore, seasonal or annual average ef-
ficiencies are used to evaluate overall system performance. Annual or
seasonal consumption of electricity required to meet the annual heating
and cooling loads (or cooling load alone) can be used to determine the
average coefficient of performance (COP)for the heat pumps and air
IX-1
-------
conditioner. From the heating and cooling loads and electricity con-
sumption calculated in Chapter VIII, the annual average COPs are as
follows:
o Residence 1: air conditioner; COP = 2.09
o Residence 2: heat pump; COP =2.22
o Residence 3: heat pump; COP = 2.22.
For the 26-MW fuel-cell power plants and the combined cycle power
plant, we assume the annual average load factor to be 0.35. However,
the variation in load that results in this average cannot be readily
determined. For example, the power plant could be operating at full
load 35% of the year and be shut off for the remaining 65%. An infinite
number of variations in load could average to 35%. However, because the
plants are intended for intermediate cycling duty, they will operate at
less than the rated load for a substantial period of time. Moreover,
they will probably be shut down for a number of hours each day during
nonpeak hours. Therefore, the average load factor during operation will
be somewhere between 0.35 and 1.0. Based on the heat rates and load
factors given in Chapter V, we have chosen the following nominal average
heat rates:
o 26-MW fuel-cell power plant (SNG) - 7,600 kJ/kWh
(7,200 Btu/kWh)
o 26 MW fuel-cell power plant (naphtha) - 7,400 kJ/kWh
(7,000 Btu/kWh)
o Combined-cycle power plant - 7,200 kJ/kWh (6,800 Btu/kWh).
Because the 100-kW fuel-cell power plant is designed to meet all
electricity requirements of the townhouse complex, it has a fairly low
load factor — 0.292 — as determined in Chapter VIII. Because the
power plant must operate continuously, and therefore have many hours of
operation during low load periods, that load factor probably represents
a true average load factor for the power plant. From the variation in
electricity and heat generation efficiencies with load factor presented
IX-2
-------
in Chapter V, the average efficiencies can be determined. The average
efficiency is 0.30 for electricity conversion, and 0.36 for heat gen-
eration, assuming an average load factor of 0.292.
Using the average energy efficiency factors presented above, the
overall system efficiencies for heating and cooling residences were
determined (see Figures IX-1 through IX-5). The figures are drawn so
that the energy supplied from one component to the next is shown beside
the line connecting the components. Any external energy requirement is
shown beside the system component that requires it, with a small arrow
indicating input to that component. The efficiency of each component is
shown within the box that represents the component.
The representation of System 5 in Figure IX-5 is slightly different
than the others. It shows 25.7 GJ of heat supplied to the space heat
delivery system from the 100-kW fuel-cell power plant. Only part of
that heat (12.8 GJ) is coproduced with the generation of electricity to
supply the heat pump. The remainder (12.9 GJ) is coproduced with the
generation of electricity for lights and appliances. However, the
figure shows only the fuel supply to the power plant needed to produce
power for the heat pump. That is because electricity is required for
lights and appliances even with no heat demand, and, after DHW demand is
met, the excess heat generated by the power plant would be rejected to
the atmosphere if there were no heating load. Because a heating load
does exist, however, heat that would have been wasted is used effec-
tively to reduce the electricity required by the heat pump. Therefore,
the energy efficiency for heating and cooling is based on the incre-
mental electricity consumption by the heat pump assuming that the light
and appliance as well as the DHW loads constitute the base system demand.
To clarify the energy supply and demand picture for System 5,
Figure IX-6 shows the annual flows of energy per residence, including
light and appliance and DHW demand.
IX-3
-------
0.5 GJ
7.3 GJ
29.4 GJ
UNIT TRAIN
1.0
29.4 GJ
COAL-FIRED
POWER PLANT
0.34
2780 kWh
ELECTRICITY
DISTRIBUTION
0.91
1901 kWh
AIR
CONDITIONER
2.09
COAL MINE
227 GJ
629 kWh
198 GJ
COAL
GASIFICATION
PLANT
0.74
147 GJ
GAS PIPELINE
0.92
135 GJ
GAS
DISTRIBUTION
1.0
135 GJ
GAS FURNACE
0.60
14.3 GJ
80.9 GJ
14.3 + 80.9
ENERGY EFFICIENCY = = 0.41
227 + 7.3 + 0.5
FIGURE IX-1. ANNUAL ENERGY FLOWS AND ENERGY EFFICIENCY
FOR SYSTEM 1
IX-4
-------
4.4 GJ
COAL MINE
i
137 GJ
COAL
GASIFICATION
PLANT
0.74
1
102 GJ
GAS PIPELINE
0.92
1
93.4 GJ
r
GAS
DISTRIBUTION
1.0
i
93.4 GJ
r
26-MW
FUEL CELL
POWER PLANT
0.47
1
12,290 k
F
ELECTRICITY
DISTRIBUTION
0.97
i
1 1 ,920 k
HEAT PUMP
2.22
95.2 GJ
ENERGY EFFICIENCY ' = 0.67
I <5 / ' 4.4
FIGURE IX-2. ANNUAL ENERGY FLOWS AND ENERGY
EFFICIENCY FOR SYSTEM 2
IX-5
-------
4.5 GJ
0.5
0.1
COAL MINE
142 GJ
COAL
LIQUEFACTION
PLANT
0.64
90.9 GJ
LIQUIDS
PIPELINE
1.0
90.9 GJ
NAPHTHA
DISTRIBUTION
1.0
90.9 GJ
26-MW
FUEL CELL
POWER PLANT
0.49
12,290 kWh
ELECTRICITY
DISTRIBUTION
0.97
11,920 kWh
HEAT PUMP
2.22
95.2 GJ
ENERGY EFFICIENCY
95.2
142 + 4.5 + 0.5 + 0.1
0.65
FIGURE IX-3. ANNUAL ENERGY FLOWS AND ENERGY EFFICIENCY
FOR SYSTEM 3
IX-6
-------
4.6 GJ
0.6
0.1
COAL MINE
i
143 GJ
r
COAL
LIQUEFACTION
PLANT
0.66
i
94.3 GJ
LIQUIDS
PIPELINE
1.0
i
94.3 GJ
r
FUEL OIL
DISTRIBUTION
1.0
i
94.3 GJ
COMBINED
CYCLE
POWER PLANT
0.50
i
13.100 k
ELECTRICITY
DISTRIBUTION
0.91
i
11.920 k
1
HEAT PUMP
2.22
95.2 GJ
ENERGY EFFICIENCY
95.2
143 + 4.6 + 0.6 + 0.1
0.64
FIGURE IX-4. ANNUAL ENERGY FLOWS AND ENERGY EFFICIENCY
FOR SYSTEM 4
IX-7
-------
2.2 GJ
COAL MINE
69.9 GJ
COAL
GASIFICATION
PLANT
0.74
51.7 GJ
GAS PIPELINE
0.92
47.6 GJ
GAS
DISTRIBUTION
1.0
47.6 GJ
100-kW
FUEL CELL
POWER PLANT
0.30
3967 kWh
25.7 GJ
HEAT PUMP
2.22
HEAT
DELIVERY
0.88
31.7 GJ
22.7 GJ
ENERGY EFFICIENCY = 31'7 + 22-7 = Q ?5
69.9 + 2.2
FIGURE IX-5. ANNUAL ENERGY FLOWS AND ENERGY
EFFICIENCY FOR SYSTEM 5
IX-8
-------
ELECTRICITY
(L & A)
31.8 GJ •*•
ELECTRICITY
14.3 GJ
HEAT PUMP
SNG
153.4 GJ
WASTE HEAT
52.2 GJ
100 kW
FUEL CELL
POWER PLANT
UNUSED HEAT
16.2 GJ
RECOVERED
HEAT
55.2 GJ
HEAT
DELIVERY
DHW
16.3 GJ
31.7 GJ
SPACE HEATING
AND COOLING
22.7 GJ
SPACE HEATING
FIGURE IX-6. TOTAL ANNUAL ENERGY FLOWS (PER RESIDENCE) FOR FUEL
CELL POWER PLANT SUPPLYING TOWNHOUSES
IX-9
-------
Based on heating and cooling only, System 5 has the highest overall
energy efficiency — 0.75. System 1 has the lowest efficiency at 0.41.
Systems 2, 3, and 4 are comparable, although System 2 has a slight
advantage at 0.67.
Overall, Systems 2 through 5 are considerably more efficient
(greater than 50%) than System 1, primarily because of the high effici-
encies of the advanced electricity generating technologies and the use
of heat pumps rather than the gas furnace. The advantages in recovering
fuel-cell heat in System 5 are clearly demonstrated, although they are
not dramatic; the efficiency is just 12% greater than System 2, the next
most efficient system. Even though heat is recovered from the 100-kW
phosphoric acid fuel cell, the electricity generating efficiency of the
26-MW molten carbonate fuel cell is 50% higher. This high efficiency,
combined with an efficient heat pump to generate space heating, consi-
derably reduces the advantage of recovering heat from a low electrical
efficiency phosphoric acid fuel cell. Optimizing of the relative
amounts of heat and electricity generated by the 100-kW fuel cell (de-
termined by the actual heat and electrical loads it is expected to meet)
would result in higher system efficiency. However, such an optimization
was beyond the scope of this study.
B. Economics
1. Cost of Heating and Cooling
The cost of providing heating and cooling to the residences
can be readily calculated using cost factors developed in Chapter VII
and the energy consumption estimates derived in Chapter VIII. The
figure of merit to be used in comparing the five systems is the cost of
providing 1 GJ of heating and cooling, which is derived by dividing the
annual cost of providing heating and cooling by the total annual heating
and cooling load. The annual cost of heating and cooling is the sum of
the fixed annualized cost of heating and cooling equipment plus the cost
of gas and/or electricity consumed by the equipment.
IX-10
-------
The costs derived by this method represent the true incre-
mental cost to the consumer of obtaining heating and cooling from each
of the five systems. These costs would never appear on the gas or
electric bill of a customer as shown here, of course, because new
sources of gas or electricity supply are integrated into the supply
system, and the customer sees only the average cost of supplying gas or
electricity from all sources in the system.
The costs of the various system components derived in Chapter
VII may be used directly as shown, except that the cost of electricity
transmission and distribution must be derived using Equations VII-(5)
and VII-(6), and the cost of electricity from the 100-kW fuel-cell power
plant is somewhat different than that shown in Table VIII-34 because of
differing load factors and because recovered heat is not explicitly
credited.
The costs of electricity transmission and distribution for
each of the residences are based on the total electricity consumption
shown in Chapter VIII and calculated using Equations VII-(5) and
VII-(6). Those costs are as follows:
o Residence 1 - 16.5 mills/kWh (System 1)
o Residence 2 - 12.7 mill/kWh (Systems 2 and 3)
o Residence 3 - 14.0 mills/kWh (System 4).
The annual average costs of heating and cooling for each of
the five systems are shown in Tables IX-1 through IX-5. The cost shown
next to each system component is the cumulative cost of energy supplied
by the component. The heating and cooling cost calculation is
summarized at the bottom of each table.
Not surprisingly, System 1 has the lowest cost of heating and
cooling because it employs the most conventional technology and has the
fewest energy conversion steps. Furthermore, even though it is the
least efficient system, the cost of the coal that the system uses is so
IX-11
-------
low ($0.33/GJ) that the effect of low efficiency on the final cost of
heating and cooling is not significant. (The cost of coal contributes
only $0.79 to the total heating and cooling cost of $10.10/GJ).
Table IX-1
COST OF HEATING AND COOLING FOR SYSTEM 1
Coal Mine $0.33/GJ
Unit train $0.61/GJ
Coal-fired power plant $9.81/GJ
(35.3 mills/kWh)
Electricity transmission $15.40/GJ
and distribution (55.3 mills/kWh)
Coal gasification plant $2.90/GJ
Gas pipeline $3.54/GJ
Gas distribution $4.16/GJ
Air Conditioner/gas furnace $294/yr
Cost of heating = $294 + 1,901 kWh/($0.0553/kWh) + 135 GJ ($4.16/GJ)
and cooling 95.2 GJ
= $10.1/GJ
Table IX-2
COST OF HEATING AND COOLING FOR SYSTEM 2
Coal mine $0.33/GJ
Coal gasification plant $2.90/GJ
Gas pipeline $3.54/GJ
Gas distribution $3.98/GJ
26-MW fuel-cell power plant $16.40/GJ
(59.0 mills/kWh)
Electricity distribution $20.40/GJ
(73.5 mills/kWh)
Heat pump $351/yr
Cost of Heating and Cooling = $351 + 11,920 kWh ($0.0735/kWh)
95.2 GJ
= $12.90/GJ
IX-12
-------
Table IX-3
COST OF HEATING AND COOLING FOR SYSTEM 3
Coal mine $0.33/GJ
Coal liquefaction plant $3.77/GJ
Liquids pipeline $3.93/GJ
Naphtha distribution $4.01/GJ
26-MW fuel-cell power plant $16.70/GJ
(60.1 mills/kWh)
Electricity distribution $20.80/GJ
(74.7 mills/kWh)
Heat pump $351/yr
Cost of Heating and Cooling = $351 + 11.920 kWh ($0.0747/kWh)
95.2 GJ
= $13.00/GJ
Table IX-4
COST OF HEATING AND COOLING FOR SYSTEM 4
Coal mine $0.33/GJ
Coal liquefaction plant $3.02/GJ
Liquids pipeline $3.17/GJ
Fuel oil distribution $3.29/GJ
Combined-cycle power plant $13.00/GJ
(46.9 mills/kWh)
Electricity transmission and distribution $18.20/GJ
(65.5 mills/kWh)
Heat Pump $351/yr
Cost of Heating and Cooling = $351 + 11,920 kWh ($0.0655/kWh)
95.2 GJ
= $11.90/GJ
IX-13
-------
Table IX-5
COST OF HEATING AND COOLING FOR SYSTEM 5
Coal mine $0.33/GJ
Coal gasification plant $2.90/GJ
Gas pipeline $3.54/GJ
Gas distribution $4.16/GJ
100-kW fuel-cell power plant $24.2/GJ
(87.1 mills/kWh)
Heat pump $314/yr
Heat delivery system $271/yr
Cost of Heating and Cooling* = $314 + $271 + 3,970 kWh ($0.0871/kWh)
70.7 GJ
= $13.20/GJ
* Includes DHW
The heating and cooling costs of Systems 2 and 3 are
comparable primarily because the cost of delivered fuels for the power
plants are nearly identical. Although coal-derived naphtha is much more
expensive to produce than SNG, the transport and distribution costs of
SNG are much higher than for naphtha.
The heating and cooling cost for System 4 is about 8% lower
than for Systems 2 and 3, primarily because of the lower fuel and
capital costs for the combined-cycle power plant. That difference is
partly offset by higher T&D costs and transmission losses, but not
enough to raise the final heating and cooling cost to the level of those
of the fuel-cell systems.
The cost of heating and cooling for System 5 also includes the
costs of supplying DHW because the heat recovery and delivery systems
are designed to provide both space heating and hot water, and their
costs cannot be readily separated. On this basis, the cost of heating
and cooling for System 5 are marginally higher than for Systems 2 and 3,
although that result is, of course, sensitive to the nature of loads
supplied by the fuel-cell power plant.
IX-14
-------
To determine the effects of load variations on the cost of
heating and cooling, those costs were determined for heating loads both
higher and lower than the Residence 3 loads derived in Chapter VIII.
The cooling load was allowed to vary with the heating load in a linear
fashion, and all other variables were held constant, including DHW load,
light and appliances loads, and power plant electrical load factor.
Holding the electrical load factor means, in effect, that the number of
residences supplied by the power plant will vary inversely with the
heating load. The annualized cost of the heat pump was allowed to vary
with the heating load while the cost of the heat delivery system, per
residence, was held constant. Finally, the quantity of heat supplied by
the fuel cell per kWh, along with the heat pump performance character-
istics, were assumed to be the same as in the Residence 3 base case.
The results of the sensitivity calculations are shown in
Figure IX-7. The cost of heating and cooling is displayed as a function
of both the annual heating load and the ratio of heating load to light
and appliance load. The light and appliance load constitutes the base
electrical demand, which determines how much of the heat is available
for space heating and DHW.
Figure IX-7 clearly shows the effect of increasing heating
load on the system economics. In particular, a heating load equal to
that of Residence 2 (80.9 GJ per year) results in a heating and cooling
cost of $11.50/GJ, which is less than Systems 2, 3, or 4. Overall, the
cost of heating and cooling varies from +17% to -14% of the base case
over a range of 0.5 to 2.0 times the base case heating load.
One difficulty in comparing the heating and cooling costs of
System 5 with those of the other systems is that the fuel-cell power
plant in System 5 supplies all electricity for the residences. In
Systems 1 though 4, the light and appliances loads are presumed to be
supplied by grid electricity (and some gas in System 1), which costs
about $0.04/kWh in 1977 prices, while the light and appliance loads in
System 5 are supplied by expensive electricity from the fuel-cell power
IX-15
-------
20
18
RATIO OF HEATING LOAD TO LIGHT AND APPLIANCE LOAD - MJ/kwh
4 6 8 10 12 14
16
18
16
8
o
Q
<
CD 14
Z
111
I
12
8
10
t
I
\
\
I
I
10
20
30 40 50 60
HEATING LOAD - GJ
70
80
90
FIGURE IX-7. VARIATION IN THE COST OF HEATING AND COOLING WITH HEATING LOAD - SYSTEM 5
-------
plant. To provide a more uniform comparison of System 5 with the other
systems, the total annual energy costs of Residence 3 were determined,
as supplied by Systems 1 through 5.
To calculate the total annual energy costs for Systems 2
through 4 supplying Residence 3, the light and appliance load for
Residence 3 was increased from 8,820 to 13,350 kWh per year to account
for DHW demand, and the heat pump load was increased from 3,967 to
6,967 kWh per year to account for the fact that recovered fuel-cell heat
would not be available. For System 1, the light and appliance load of
Residence 3 was reduced by 1,200 kWh per year because of the absence of
the electric range. Gas consumption by appliances included 32.9 GJ per
year for DHW and 11.1 GJ for the gas range. Furthermore, the figures
for consumption of gas and electricity for heating and cooling were
adjusted to account for the lower loads of Residence 3 relative to
Residence 1.
The average grid price of electricity for Systems 1 through 4
was determined from total electrical loads and Equations VII-(4), (5),
and (6). The average gas price for supplying DHW and range loads for
System 1 was assumed to be the 1977 average of about $2.00/GJ. The
annualized cost of heating and cooling equipment was reduced to account
for the lower heating and cooling loads of Residence 3.
With the assumptions discussed above, the following total
annual energy supply costs for Residence 3 as supplied by Systems 1
through 5 were calculated:
o System 1 cost = $250 + 75 GJ ($4.16/GJ)
+ 44 GJ ($2.00/GJ) + 1,605 kWh (0.0553/kWh)
+ 7,620 kWh ($0.0391/kWh)
= $1,040
o System 2 cost = $314 + 13,348 kWh ($0.0366/kWh)
+ 6967 kWh ($0.0739/kWh)
= $1,320
IX-17
-------
o System 3 cost = $314 + 13,348 kWh ($0.0366/kWh)
+ 6,967 kWh (0.0751/kWh)
= $1,330
o System 4 cost = $314 + 13,348 kWh ($0.0366/kWh)
+ 6,967 kWh ($0.0661/kWh)
= $1.270
o System 5 cost = $314 + $271 + 12,787 kWh ($0.0871/kWh)
= $1,700
System 5 is not economical in terms of supplying the total
energy requirements of Residence 3 compared with other systems. The
high cost of System 5 is partly caused by the low power plant load
factor, partly by the high cost of the heat delivery system, and partly
by the use of the fuel-cell power plant to supply electricity to the
residences at all times, even when the heat demand is very low.
The optimum arrangement for System 5 would be to use grid
electricity during periods of low heat demand and to use electricity
from the fuel-cell power plant when the heat demand is high. The
appropriate mix would have to be determined through an optimization
procedure that is beyond the scope of this study.
Moreover, the 100-kW power plant in System 5 cannot be
expected to meet any conceivable load by itself, even if it is designed
to meet the largest average load in a typical year. Excursions above
the power plant rated load could readily occur on cold winter days.
Also, during extreme temperature periods, the total load could be well
in excess of design capacity. The only feasible solutions are for the
residences to be connected to the utility grid as a backup in high
demand periods, or to have a load management system in which noncritical
appliances (such as clothes dryers) would be automatically shut off
during high demand periods. In the former case, the residences would
have to pay a utility hook-up charge of about $3.57 per month (according
to Equation VII-(4)) even when no power was consumed, which would add
about $0.19/GJ to the cost of heating and cooling or $42.80 per year to
the total annual residential energy cost. An additional cost would also
be incurred for a load management system, although such a cost was not
determined in this study.
IX-18
-------
2. Capital Intensiveness
Another measure of the relative economic attractiveness of the
five systems is the amount of capital required to install the various
system components. Because capital may be considered a scarce resource,
the capital intensiveness of the systems will measure the relative ease
of initially establishing the systems, which is independent of the
life-cycle system cost that was calculated in the previous section.
The unit appropriate for measuring the capital intensiveness
of Systems 1-5 is the capital required to provide 1 GJ of heating and
cooling per year, which may be calculated by using the component energy
flows displayed in Figures IX-1 through IX-5, along with the capital
intensiveness for each system component. However, the calculation of
component capital intensiveness is complicated somewhat; two figures may
be derived: (1) the amount of capital investment required per unit of
peak energy output, and (2) the capital required per unit of average
energy output. For the systems under consideration in this study, the
capital investment per unit of average energy output will be used,
because, by and large, the systems components are not sized to meet peak
system demands and are effectively decoupled from one another. For
example, the output from the coal gasification plant is fairly constant
in time, while the output of the fuel-cell power plant that it supplies
fluctuates daily and even hourly. The inherent storage capacity of the
natural gas transport and distribution system, plus the varying demands
of other users of SNG, effectively eliminate such demand fluctuations in
the coal gasification plant output, and peak demands have relative
little effect on the plant's capacity. That argument is not strictly
true for each system component — for example, the 100-kW fuel-cell
power plant is sized to meet the peak electricity demand of the
townhouse in an average year. However, because the argument is
generally applicable, the average energy outputs will be used for all
systems components for consistency.
IX-19
-------
The resulting figures for capital intensiveness of the system
components are shown in Table IX-6. Those figures are based on the
total capital investments presented in Chapter VII and the yearly energy
output of the system components. In some cases, capital investments
were not explicitly given in Chapter VII, and the capital intensiveness
was estimated using the data at hand. For example, the capital inten-
siveness of gas distribution was estimated by assuming that 90% of gas
distribution charges are capital-related (the same proportion as for the
gas pipeline) and by using a capital recovery factor of 18% per year
(typical for utility economics). The figure for the coal-fired power
plant represents one-half of the initial investment because the plant is
assumed to be 50% depreciated when it is assigned to intermediate-load
duty.
Table IX-6
CAPITAL INTENSIVENESS OF THE SYSTEM COMPONENTS
Component Capital Intensiveness
Coal mine $0.41 per GJ/yr
Unit train $0.53 per GJ/yr
Coal-fired power plant $1,210 per kW (avg.)
Coal gasification plant $11.40 per GJ/yr
Coal liquefaction plant (fuel oil) $11.40 per GJ/yr
Coal liquefaction plant (naphtha) $12.60 per GJ/yr
Gas pipeline $ 2.20 per GJ/yr
Liquids pipeline $ 0.74 per GJ/yr
Liquid fuel distribution (fuel oil) $ 0.09 per GJ/yr
Liquid fuel distribution (naphtha) $ 0.11 per GJ/yr
Gas distribution
Residential $3.10 per GJ/yr
Commercial $2.20 per GJ/yr
Combined-cycle power plant $911 per kW (avg.)
26-^IW fuel-cell power plant (SNG) $1,180 per kW (avg.)
26-MW fuel-cell power plant (naphtha) $1,240 per kW (avg.)
Electricity transmission and
distribution
Central power plant $550 per kW (avg.)
Dispersed power plant $490 per kW (avg.)
100-kW fuel-cell power plant $1,570 per kW (avg.)
Gas furnace $11.00 per GJ/yr
Air conditioner $86.60 per GJ/yr
26-MJ/hr heat pump $26.50 per GJ/yr
Residence 3 heating and cooling system $71.90 per GJ/yr
IX-20
-------
The figures for capital intensiveness presented in Table IX-6
were combined with the energy flows shown in Figures IX-1 through IX-5
to obtain the total capital intensiveness per GJ/yr of heating and
cooling for each of the five systems (see Tables IX-7 through IX-11).
The trends are similar to those displayed in the cost of heating and
cooling calculations shown in Tables IX-1 through IX-5. The capital
intensiveness of System 1 is the lowest of the five, as expected.
Systems 2 through 4 are comparable, with System 4 having about a 10%
advantage over system 2. The capital intensiveness of System 5 is much
higher than that of any other system, primarily because of the high
investment required for the heat delivery system. If heating and
cooling equipment, the cost of which is borne by the consumer rather
than the utilities, is excluded from the calculations, System 5 has the
lowest capital intensiveness of any system. Therefore, System 5 would
be very attractive to the utilities because of the initial investment
Table IX-7
CAPITAL INTENSIVENESS FOR SYSTEM 1
($ per GJ/yr of Heating and Cooling)
Coal mine $0.98
Unit train 0.17
Coal-fired power plant 4.02
Electricity transmission
and distribution
Air conditioner
Coal gasification plant
Gas pipeline
Gas distribution
Gas furnace
Total
IX-21
-------
Table IX-8
CAPITAL INTENSIVENESS FOR SYSTEM 2
($ per GJ/yr of Heating and Cooling)
Coal mine $ 0.59
Coal gasification plant 12.10
Gas pipeline 2.18
Gas distribution 2.16
26-MW fuel-cell power plant 17.30
Electricity transmission
and distribution 7.04
Heat pump 26.50
Total $67.87
Table IX-9
CAPITAL INTENSIVENESS FOR SYSTEM 3
($ per GJ/yr of Heating and Cooling)
Coal mine $ 0.61
Coal liquefaction plant 12.00
Liquids pipeline 0.71
Naphtha distribution 0.10
26-MW fuel-cell power plant 18.20
Electricity transmission
and distribution 7.04
Heat pump 26.50
Total $65.10
IX-2 2
-------
Table IX-10
CAPITAL INTENSIVENESS FOR SYSTEM 4
($ per GJ/yr of Heating and Cooling)
Coal mine $ 0.62
Coal liquefaction plant 11.30
Liquids pipeline 0.73
Fuel oil distribution 0.09
Combined-cycle power plant 14.30
Electricity transmission
and distribution 7.87
Heat pump 26.50
Total $61.40
Table IX-11
CAPITAL INTENSIVENESS FOR SYSTEM 5
($ per GJ/yr of Heating and Cooling)
Coal mine 0.41
Coal gasification plant 8.33
Gas pipeline 1.49
Gas distribution 2.09
100-kW fuel-cell power plant 10.10
Heat pump/heat delivery 71.90
Total 94.30*
Includes DHW for this system only.
IX-23
-------
required. However, the consumer would be discouraged from participating
in such an arrangement because of the large initial equipment cost.
C. Environmental Impact
The environmental aspects of the system components were analyzed in
Chapter VI. Those analyses can be used to develop enviromental impact
profiles of the five systems. Qualitative judgments play an important
role in comparing the systems because an absolute quantitative ranking
of the systems is generally not possible, nor would it necessarily be
desirable.
1. Pollutant Emissions
The emissions of air and water pollutants and solid wastes
developed in Chapter VI for system components were used to calculate the
total emissions for the five systems. So that the systems could be
compared on an equivalent basis, the emissions were normalized on the
basis of pollutants emitted per GJ of heating and cooling. These
quantities are shown in Tables IX-12 through IX-16. Because the unit of
reference is 1 GJ of heating and cooling, the amount of energy issuing
from each component is generally greater or less than that amount, de-
pending on the various component efficiencies, and is shown in
parentheses below the name of each system component.
A dash in any column means that none (or an insignificant amount)
of that pollutant is produced. The entries in each pollutant emission
column have not been added to obtain total emissions per GJ of heating
and cooling, primarily, because emissions from the various system
components take place in different geographical regions and the
resulting pollutant burden to the environment must be examined in each
location. Also, point sources (e.g., coal gasification plant) occur
over a relatively small geographical area, while other emissions are
spread over a large area (e.g., unit trains). The resulting pollutant
burdens to the environment are considerably different depending on the
nature of the source.
IX-24
-------
Ul
Table IX-12
POLLUTANT EMISSIONS ASSOCIATED WITH SYSTEM 1
(g Per GJ of Heating and Cooling)
Coal Mine
(2.38)*
Air Pollutants
S02 0.38
HOX 6.4
Part.** 0.21/19
HC 0.36
CO 1.1
PAH 1.8 x 10-5
Sb —
As —
Be —
Cd —
Pb —
tt« — —
**g
Se —
Zn —
Water Pollutants
Suspended Solid 0.76
Oil and Grease —
Iron 0.013
Manganese 0.0026
Copper —
Chlorine —
Solid Waste —
Electricity
Unit Train Power Plant T&D
(0.30)* (0.102)* (0.093)*
0.80 28
5.0 91
0.36/_ 1.8/4.2
1.3 2.2
1.8 7.3 —
2.2 x 10"5 8.4 x 10-5
~ 2.2 x 10-*
1.2 x 10-5
8.1 x 10-5 _
6.0 x 10-*
4.6 x 10-3
1.4 x 10-3
7.1 x 10-4
7.1 x 10-3
0.71
— 0.35
0.024
_ _ —
0.024 —
4.6 x 10-4
— 1,900 —
Air Coal Gasifi- Gas
Conditioner cation Plant Pipeline
(0.15)* (1.54)* (1.42)*
33 0.033
•75 17
— / 3 Li
4.3/18 0.85/-
4.9 1.4
6.3 6.8
3.8 x 10-* 5.4 x ID"5
1.4 x 10-*
8.* x 10-6
— 5.0 x 10-5
3.8 x 10-*
2.9 x 10-3
8.8 x 10-*
4.5 x 10-*
4.5 x 10-3
__ —
—
13,000
Gas Gas
Distribution Furnace
(1.42)* (0.85)*
0.36
49
6.1/.
— 4.9
12
6.2 x 10~4
— —
—
—
—
— —
--
—
—
GJ supplied by the system component per GJ of heating and cooling.
**Fine particulates/coarae participates.
-------
Table IX-13
POLLUTANT EMISSIONS ASSOCIATED WITH SYSTEM 2
(g Per GJ of Heating and Cooling)
Air Pollutants
S02
NOX
Part.**
HC
CO
PAR
Sb
As
Be
Cd
Pb
Hg
Se
Zn
Water Pollutants
Suspended Solid
Iron
Manganese
Solid Waste
Coal Mine
(1.44)*
0.23
3.9
0.13/11
0.22
0.67
1.1 x 10-5
-
-
-
—
—
-
—
-
0.46
0.0076
0.0016
—
Coal Gasifi- Gas Gas Fuel-Cell Electricity
cation Plant Pipeline Distribution Power Plant Distribution Heat Pump
(1.06)* (0.980)* (0.980)* (0.464)* (0.450)* (1.0)*
23 0.022
52 12 — 5.3 x 10-7
3.0/12 0.59/-
3.4 0.97
4.3 4.7
2.6 x 10-* 3.7 x 10-5
9.6 x 10-5
5.8 x 10-6
3.4 x 10-5
2.6 x 10-4
2.0 x 10-3
6.1 x 10-*
3.1 x 10-*
3.1 x 10-3
_
-
-
8,900
*GJ supplied by the system component per GJ of heating and cooling.
**Fine particulates/coarse particulates
IX-26
-------
N>
Table IX-14
POLLUTANT EMISSIONS ASSOCIATED WITH SYSTEM 3
(g per GJ of Heating and Cooling)
Coal Mine
(1.49)*
Air Pollutants
S02 o.24
»°x 4.0
Part.** 0.13/12
BC 0.23
CO 0.69
PAH 1.1 x 10-5
Sb —
Aa
Be
Cd —
Pb —
Hg
Se
Zn
Water Pollutants
Suspended Solid 0.48
Iron 0.0079
Manganese 0.0016
Solid Haste —
Coal
Liquefaction Liquids Naphtha Fuel-Cell Electricity
Plant Pipeline Distribution Power Plant Distribution Heat Pump
(0.954)* (0.954)* (0.954)* (0.464)* (0.450)* (1.0)*
29 0.55 0.11
95 8.4 0.71 1.1 x 10-5 — _
1.9/21 0.61/_ 0.052/.
2.1 0.67 0.19
7.3 1,8 1.2 — — —
1.5 x 10-4 2.8 x 10-5 6.i x ifl-6
2.0 x 10-4 — — — — _-
1.1 x 10-5
7.5 x 10-5 _
5.5 x 10-4 _ _. _ „ _
4.2 x 10-3 _
1.3 x 10-3
6.4 x 10-4
6.4 x 10-3 _
_
—
—
8,800
*GJ supplied by the system component per GJ of heating and cooling.
**Fine particulates/coarse particulates.
-------
Table IX-15
POLLUTANT EMISSIONS ASSOCIATED WITH SYSTEM 4
(g per GJ of Heating and Cooling)
Coal Mine
(1.50)*
Air Pollutants
S02 0.24
NOX 4.0
Part.** 0.13/12
HC 0.23
CO 0.69
PAH 1.1 x 10-5
Sb
As
Be
Cd
Pb —
Hg
Se
Zn
Water Pollutants
Suspended Solid 0.48
Oil and Grease —
Iron 0.0080
Manganese 0.0017
Copper
Chlorine
Solid Waste
Coal
Plant Pipeline
(0.990)* (0.990)*
29 0.57
95 8.7
1.9/21 0.63/.
2.1 0.70
7.3 1.9
1.5 x 10-4 2.9 x 10-5
2.0 x 10-4
1.1 x 10-5
7.5 x 10-5
5.5 x 10-4
4.2 x 10-3
1.3 x 10-3
6.4 x 10-*
6.4 x 10-3
—
-
—
-
-
—
8,800
Combined-
Distribution Power Plant T & D Heat Pump
(0.990)* (0.495)* (0.450)* (1.0)*
0.32 46
2.1 122
0.15/. 24/_
0.55 12
0.76 12
4.9 x 10-6 2.3 x 10-3
1.5 x 10-3
6.9 x 10-3
1.6 x 10-3
4.8 x 10-3
0.016
2.3 x 10-*
1.7 x 10-3
0.076
2.9
— 1.4
0.10
„
0.10
1.5 x 10-3
—
*GJ supplied by the system component per GJ of heating and cooling.
IX-28
-------
Table IX-16
POLLUTANT EMISSIONS ASSOCIATED WITH SYSTEM 5
(g per GJ of Heating and Cooling)
Coal Mine
(1.28)*
Air Pollutants
S02 0.20
NO, 3.4
Part.** 0.11/10
HC 0.19
CO 0.59
PAH 9.7 x 10-6
Sb —
As
Be
Cd —
Pb
Hg
Se
Zn
Water Pollutants
Suspended Solid 0.41
Iron 0.0068
Manganese 0.0014
Solid Waste
Coal
Gasification Gas
Plant Pipeline
(0.951)* (0.875)*
20 0.02
46 10
2.7/11 0.52/_
3.0 0.86
3.9 4.2
2.3 x 10-4 3.3 x io-5
8.6 x 10-5
5.2 x 10-6
3.1 x 10-5
2.3 x 10-*
1.8 x 10-3 —
5.4 x 10-4 ~
2.8 x 10-4
2.8 x 10-3
__ «
__
__
8,000
Heat Pump
Gas Fuel-Cell and
Distribution Power Plant Heat Recovery
(0.875)* (0.978)* (1.0)*
—
6.3
—
4.2
—
—
__
__
__
—
__
__
—
__
__ __
__
__
__ — — —
**,
*GJ supplied by the system component per GJ of heating and cooling.
'Pine particulates/coarse particulates.
IX-29
-------
To account for such factors, the pollutant emissions from the
systems components were classified into three categories — (1) those
emitted in the mine or near-mine region, (2) those emitted during
transport to the end-use region, and (3) those emitted in the end-use
region. The resulting assignment of system components to each category
is shown below:
o Category 1 - coal mine, coal gasification plant, coal
liquefaction plant
o Category 2 - unit train, gas pipeline, liquids pipeline
o Category 3 - coal-fired power plant, combined-cycle power
plant, fuel-cell power plants, liquid fuel
delivery, gas furnace
Although the emission of pollutants in each category can be
considered separately, an overall emission parameter for each pollutant
for an entire system is desirable. Such parameters can be formulated if
appropriate weighting factors can be assigned for pollutant emissions in
each category. Weighting factors should be chosen on the basis of the
likely effects of pollutant emissions on human health and the
environment in each category.
First, because the emissions from Category 2 are dispersed, they
should be weighted relatively less than the emissions from Categories 1
or 3. The gas and liquids pipelines have 10 or 11 separate emission
sources spaced at about 80 km (50 mi), whereas the unit train emissions
are continuous. Because emissions from different pipeline pumping
stations are unlikely to interact, the total emissions can be assigned a
weighting factor of about 1/10. This factor will also be applied to the
unit train emissions to equalize all components in Category 2.
The relative weighting of Categories 1 and 3 is complex and
difficult. On one hand, Category 3 emissions could be weighted higher
because (1) the Omaha-Kansas City-Des Moines region is much more highly
populated than the Powder River Basin, and therefore the human exposure
to a given pollutant release will be greater, (2) additional pollutant
emissions will exacerbate an existing urban pollution problem, and (3)
IX-30
-------
the highly productive agricultural environment of the region could be
adversely affected, whereas the Powder River Basin has low agricultural
productivity. On the other hand, Category 1 emissions could be weighted
higher because (1) the Powder River Basin is a near-pristine environment
that should be protected from significantly increased pollutant levels,
(2) the growth in energy production in the region will substantially
increase the population that will be exposed to pollutants, and (3) such
growth will result in pollutant levels and control problems now facing
more urbanized areas.
Because of these considerations, there is no clear-cut choice for
weighting Category 1 emissions with respect to Category 3 emissions.
Furthermore, no quantitative method can be used to assign weighting
factors with any degree of confidence. Therefore, we decided to weight
the two categories equally. That choice does not mean that the impacts
of pollutant emissions in the two regions will be the same. Rather, it
acknowledges that the differences in impacts are not resolvable within
the scope of this study.
The weighted pollutant emissions that result from application of
the weighting factors discussed above are shown in Table IX-17.
Those emissions can be used to compare the overall environmental impact
of the five systems. However, to fully account for the relative effects
of the various pollutants emitted from the systems components, some
standard that serves as a measure of those effects must be used. Such
standards, and their use in arriving at relative system rankings, are
discussed below.
a. Air Pollutants
Although ambient air quality standards have been set for
SO-, NO , CO, hydrocarbons and particulates, those for PAH and trace
*The weighted emissions are equal to (Category 1 emissions plus one-tenth of
Category 2 emissions plus Category 3 emissions) divided by 2.1.
IX-31
-------
Table IX-17
GEOGRAPHICALLY WEIGHTED POLLUTANT EMISSIONS FOR THE FIVE SYSTEMS
(g per GJ of Heating and Cooling)
System 1
System 2
System 3
System 4
System 5
Air Pollutants
S02
NOX
Part.*
HC
CO
PAH
Sb
As
Be
Cd
Pb
Hg
Se
Zn
Water Pollutants
Suspended Solid
Oil and Grease
Iron
Manganese
Copper
Chlorine
Solid Waste
29.4
106
6.9/19.6
6.0
13.1
5.3 x 10-4
1.7 x 10-5
9.7 x 10-6
6.2 x 10-5
4.7 x 10-4
3.6 x 10-3
1.1 x 10-3
5.5 x 10-4
5.5 x 10-3
0.70
0.17
0.018
1.2 x 10-3
0.011
2.2 x 10-4
7,100
11.1
27.2
1.5/11.0
1.8
2.6
1.3 x 10-*
4.6 x 10-5
2.8 x 10-6
1.6 x 10-5
1.2 x 10-*
9.5 x 10-*
2.9 x 10-*
1.5 x 10-*
1.5 x 10-3
0.22
—
0.0036
7.6 x 10-5
—
—
*,200
14.0
47.9
1.0/15.7
1.2
4.5
8.1 x 10-5
9.5 x 10-5
5.2 x 10-6
3.6 x 10-5
2.6 x 10-*
2.0 x 10-3
6.2 x 10~*
3.0 x 10-*
3.0 x 10-3
0.23
—
0.0038
7.6 x 10-*
—
—
4,200
36.0
107
12.5/15.7
7.1
10.0
1.2 x 10-3
8.1 x 10-*
3.3 x 10-3
8.0 x 10-*
2.7 x 10-3
9.6 x 10-3
7.3 x 10-*
1.1 x 10-3
3.9 x 10-2
1.6
0.67
0.051
8.1 x 10~*
0.048
7.1 x 1004
4,200
9.6
27.0
1.4/10.0
3.6
2.3
1.2 x 10-*
4.1 x 10-5
2.5 x 10-6
1.5 x 10-5
1.1 x 10-*
8.6 x 10-*
2.6 x 10~*
1.3 x 10-*
1.3 x 10-3
0.20
—
0.0032
6.7 x 10-*
—
__
3,800
*Fine particulates/coarse particulates.
IX-32
-------
elements still need to be determined. Fortunately, occupational
exposure standards have been developed by the Occupational Safety and
Health Administration (OSHA) and have been recommended by the American
Conference of Governmental Industrial Hygienists (ACGIH) and the
National Institute for Occupational Safety and Health (NIOSH) for nearly
all pollutants of interest (see Table IX-18). The standards are
designed to protect workers who are continuously exposed for 8 hours per
day, 40 hours per week. The allowable concentrations for occupational
exposure tend to be somewhat higher than those that are designed to
protect the general public. While these standards do not provide a
direct measure of the environmental impacts of pollutants, they do
provide a basis for assessing the relative effects of pollutants on
human health.
Table IX-18
OCCUPATIONAL EXPOSURE STANDARDS FOR TOXIC POLLUTANTS
TIME-WEIGHTED AVERAGES (mg/m3)
Pollutant
S02
N02
Particulates
Hydrocarbons
CO
PAH
Sb
As
Be
Cd
Pb
Hg
Se
Zn
ACGIH
Recommendation
13
9
55
0.5
0.5
0.002
0.05
0.15
0.05
0.2
5.0
OSHA
Standard
13
2.4-10
(coal dust)*
400
(naphtha)
55
0.5
0.5
0.002
0.1
0.2
0.2
5.0
NIOSH
Recommendation
350
(alkanes)
39
0.002
0.04
0.15
0.05
5.0
^Respirable fraction.
IX-33
-------
Unfortunately, no occupational standards for PAH or any compound
included in this category currently exist. However, several of those
compounds are well known carcinogens and, in particular, exposures to
benzo(a)pyrene (BaP) has been correlated with excess lung cancers in
coke-oven workers and those in similar occupations. Such correla-
tions indicate that significant effects begin to occur at levels between
o
0.00001 and 0.001 mg/m . Given that BaP is only one of a number of
PAH compounds that may be contributing to lung cancer, and that total
3
PAH emissions are of interest here, a standard of 0.001 mg/m for
total PAH provides very roughly the same level of protection as those
standards listed in Table IX-18.
The standards shown in Table IX-18 will be used to define
the relative hazards of the pollutants listed. No attempt will be made
to attain an absolute measure of the impacts of the various system com-
ponents, and indeed, the standards were not designed to be used in such
a fashion.
Using the standards shown, weighting factors representing
the relative hazards of the various pollutants were developed (see Table
IX-19). The factors are simply the inverse of the occupational exposure
standards, but because of the roughness of the measure of relative
hazard, the factors are given to only one significant figure. There are
no standards for particulates as such, just for respirable coal dust.
However, using this analog for a particulate standard is reasonable when
deriving a weighting factor. Having the same weighting factor for
SO , NO , and particulates is also reasonable, especially
considering that the National Ambient Air Quality Standards for those
pollutants are about the same, that is, annual average values of 80,
100, and 75 mg/m for S02, N02, and particulates, respectively.
The weighting factors shown in Table IX-19 were
multiplied by the geographically weighted air pollutant emissions in
Table IX-17 and divided by the sum of the weighting factors to arrive at
a hazard-weighted air pollutant emission factor for each of the five
systems (see Table IX-20). They clearly indicate that Systems 2 and 5
IX-34
-------
Table IX-19
WEIGHTING FACTORS FOR THE RELATIVE HAZARDS
OF AIR POLLUTANTS
Pollutant Weighting Factor
PAH 1,000
Be 500
Hg 20
Cd 20
Pb 6
Se 5
Sb 2
As 2
Zn 0.2
so2 o.i
N02 0.1
Particulates 0.1
CO 0.02
Hydrocarbons 0.003
Table IX-20
HAZARD-WEIGHTED AIR POLLUTANT EMISSION FACTORS
FOR THE FIVE SYSTEMS
System 1 0.0097
System 2 0.0027
System 3 0.0042
System 4 0.013
System 5 0.0026
IX-35
-------
have the least air pollution impact; Systems 1 and 4 have the highest
and System 3 lies in between.
b. Water Pollutants
The emission of water pollutants by the five systems
occurs from only two sources — the coal mine and power plant wastewater
discharge. Neither of these is a major source of toxic water
pollutants, and the pollutants that are emitted must meet EPA effluent
guidelines. To achieve a simple ranking of the systems, it is not
necessary to derive weighting factors as was done for air pollutants
because the relative ranking of the systems is the same for each
pollutant listed in Table IX-17- Examination of Table IX-17 indicates
that System 4 has the highest water pollution impact, System 1 has the
next highest, Systems 2 and 3 are comparable, and System 5 has the
lowest.
c. Solid Waste
With the exception of System 1, the only sources of solid
waste in the five systems are the coal gasification and liquefaction
plants. Even in System 1, only 13% of the solid waste originates from
the coal-fired power plant. The remainder is from coal gasification.
It would be extremely difficult to assess the relative hazard of the two
types of waste. Both contain coal ash and char, FGD solids, and sludge
from the biological oxidation ponds. If the wastes are properly
disposed of, they present little environmental hazard. The main source
of concern is the possibility of toxic materials leaching from the waste
piles into aquifers, as discussed in Chapter VI. Again, the different
types of waste cannot be distinguished in terms of their potential for
leaching. The method of disposal and the properties of the disposal
site will probably have more bearing on the likelihood of leaching than
the composition of the waste.
IX-36
-------
Thus, no weighting factors were applied to the quantities
of solid waste produced by the systems. Therefore, according to Table
IX-17, System 1 clearly has the largest solid-waste impact, System 5 has
the least, and Systems 2, 3, and 4 are identical, with a somewhat higher
impact than System 5.
2. Land Use, Noise, and Aesthetics
Several other environmental factors should be considered when
comparing the five systems. For purposes of analysis, those factors are
categorized as land use, noise, and aesthetics.
a. Land Use
To assess the effect on land use, the amount of land
occupied or disturbed by the systems to produce the energy required for
heating and cooling residences was calculated. An appropriate measure
of land use is the area occupied or disturbed multiplied by the length
of time it is effectively removed from other purposes such as agri-
culture, housing, recreation, or wildlife support and plant habitat.
For facilities such as coal conversion plants, it is the area occupied
multiplied by the lifetime of the facility. For activities such as coal
mining, it is the area disturbed multiplied by the length of time from
mining to final reclamation. To compare the five systems, all such
measures are normalized to the energy output of the facility or activity
and ultimately to the heating and cooling supplied to residences, with
2
the impact factor measured in m -year/GJ.
Systems components that have significant effects on land
occupancy or disturbance are shown in Table IX-21. The scaling factors
are based on estimates of the land occupied by energy conversion faci-
lities, land disturbance, and right-of-way quantities presented in
Chapter VI, and on the following assumptions: (1) Four years are
IX-37
-------
required for complete reclamation of land disturbed by coal mining; (2)
land disruption from pipeline construction persists for 2 years; (3) the
land disruption from electricity transmission is based on a figure of
0.5 km (0.3 mi) of new transmission lines required per megawatt of added
2
generating capacity in Missouri, Nebraska, and Iowa, and on a
right-of-way of 30 m (100 ft). The right-of-way is assumed to be
o
disturbed for 2 years, while the land occupied by the towers (50 m
per tower with towers spaced every 0.3 km) is disturbed for the
lifetime of the line.
The right-of-way factor shown in Table IX-21 is
appropriate for central generating facilities. For dispersed fuel-cell
power plants, it should be multiplied by 0.75 to reflect the reduced
transmission requirements discussed in Section VII-M.
Using the factors presented in Table IX-21 and the energy
efficiency quantities shown in Figures IX-1 through IX-5, the land use
per GJ of heating and cooling can be calculated for the five systems
(see Table IX-22).
Table IX-21
LAND USE FACTORS FOR SYSTEM COMPONENTS
Component Land Use, (m^-yr/GJ)
Coal mine 0.018
Unit train 0.0053
Coal-fired power plant 0.12
Coal gasification plant 0.023
Coal liquefaction plant 0.021
Gas pipeline 0.011
Liquids pipeline 0.0062
Combined-cycle power plant 0.045
26-MW fuel-cell power plant 0.0081
Electricity transmission 0.057
IX-38
-------
Table IX-22
TOTAL LAND USE FOR THE FIVE SYSTEMS
(m2-year/GJ of Heating and Cooling)
System 1 0.11
System 2 0.084
System 3 0.076
System 4 0.10
System 5 0.055
Systems 1 and 4 require the greatest land use, followed
by Systems 2, 3, and 5, with System 5 needing only one-half the land of
Systems 1 and 4. We have made no judgments as to the relative value of
land in the regions encompassed by the five systems. Such complex
considerations are beyond the scope of this study. In practice, the
actual siting of the components of energy sytstems, and therefore the
nature of the land that is disturbed, will result from trade-offs among
economic, environmental, and social factors that will be very
site-specific.
b. Noise
The noise characteristics of the system components as
they affect the general public were identified in Chapter VI. Although
many components are very noisy, not all of them need be heard by the
public. Large centralized facilities such as coal mines and coal
conversion plants are generally located sufficiently far from
residential, commercial, and recreational areas so that significant
noise levels are not disturbing. On the other hand, equipment such as
locomotives and fuel delivery trucks pass near or through populated
areas, and thus expose large numbers of people to high noise levels.
The main sources of obtrusive noise and their average
noise levels are summarized in Table IX-23. An urban residential
IX-39
-------
background noise level of 48 dBA is considered average (see Table
VI-17). Against such a background, transmission lines and fuel-cell
power plants are relatively modest sources of additional noise. In
quiet suburban residential areas with background noise levels of 38 dBA,
their noise levels would be much more noticeable and could result in
some speech masking and sleep disturbance. The noise emitted from
trains and tank trucks, however, would obviously be the most
objectionable, because it is many times higher than background levels,
even in the noisiest urban areas. However, transmission-line and
fuel-cell power plant noise is more or less continuous, but that from
trains and trucks is intermittent.
The most useful way to compare the noise impact of the
five systems is to determine the number of major noise sources contained
in each system. Thus, System 5 has the lowest impact because it has
only one low-level noise source (the 100-kW fuel-cell power plant).
System 2 has two low-level noise sources (transmission line and
fuel-cell power plant) and therefore has the next highest impact.
Systems 1 and 4 both have one high-level and one low-level noise
source. Finally, System 3, with one high-level and two low-level noise
sources, has the greatest impact, although it is only marginally greater
than Systems 1 and 4.
Table IX-23
SOURCES OF INVOLUNTARY EXPOSURE TO HIGH NOISE LEVELS
Source Noise Level (dBA)
Transmission lines 55
Fuel-cell power plants 55
Unit trains 90-100
Tank trucks 88
IX-40
-------
c. Aesthetics
Evaluation of the aesthetic impacts of the five systems,
even on a relative basis, is difficult because of the many value judg-
ments involved and the lack of quantitative bases of comparison. Fur-
thermore, the systems and their components are so similar that, with few
exceptions, broad aesthetic differences among the systems are not
readily discernible.
One noticeable aesthetic impact, however, is the visible
plume caused by the emission of coarse particulates from the operation
of some system components. (See Tables IX-12 through IX-16.) Although
such emissions have relatively little health impact, they degrade visi-
bility near the sites of system components that emit coarse particu-
lates. Such degradation constitutes a major aesthetic impact.
The relative impact of coarse particulate emissions from
each system can be evaluated by consulting Table IX-17, which presents
the geographically weighted emissions of air pollutants from each
system. According to that table, System 5 has the lowest impact, with
emissions lower by a factor of two than System 1, which has the highest
impact. System 2 has slightly higher emissions than System 5, and the
emissions of Systems 3 and 4 are equal, both being about midway between
Systems 1 and 5.
Another significant aesthetic impact of the five systems
is caused by electrical transmission lines, which are perhaps the most
extensive and noticeable aspect of the entire electrical energy system,
including its associated fuel cycle. The use of dispersed power plants
in Systems 2, 3, and 5 significantly reduces this impact. In principle,
System 5 can avoid the use of transmission lines entirely, but in prac-
tice the residences may have to be connected to the electrical grid to
ensure reliability; such a connection implies the use of some trans-
mission facilities.
IX-41
-------
The use of dispersed 26-MW power plants does not
eliminate electrical transmission entirely because interties with the
rest of the utility system are needed. As discussed in Chapter VII,
however, the requirement for transmission facilities can be reduced on
the order of 25% compared to centralized generating facilities as
represented by Systems 1 and 5.
Finally, siting dispersed fuel-cell power plants in
urban/residential areas has aesthetic implications. Such plants will be
a departure from the usual mix of homes, apartment buildings, commercial
buildings, shopping complexes, schools, parks, and so on. However, the
plants will be fairly unobtrusive, consisting of clusters of rectangular
structures about 3 m (10 ft) in height. The use of proper landscaping
and site design should mitigate any unattractive features.
D. System Performance
All the systems analyzed in this study are based on advanced coal
conversion and/or electricity generation technologies that have yet to
be proven in commercial-scale operation. The cost, efficiency, and
environmental analyses of the systems are based on the assumption that
the performance goals set by the developers of the technologies would be
achieved in practice, and that they would be capable of performing as
specified in the applications set forth in Chapter IV. The implication
of those assumptions is that the systems would be less efficient, more
costly, and more environmentally intrusive than our analyses have shown
if those performance goals were not met.
In addition to those obvious effects, the desirability of
implementing the systems will be strongly affected by considerations of
reliability, lifetime, and performance characteristics of the major
components. Thus, it is reasonable to ask what effect such
considerations will have on the relative attractiveness of each of the
five systems, and to what extent uncertainties about those
characteristics will affect the implementation of the systems.
IX-42
-------
In examining those aspects of the systems, we need address only
those components that represent truly new technologies. Other
components, such as pipelines, transmission lines, and coal mines, have
well known characteristics, and their performance parameters have been
established by many years of use. Therefore, little additional light
can be shed on their contribution to the overall effectiveness of the
systems.
All the systems contain either an advanced coal gasification or
coal liquefaction facility (Hygas and H-Coal) which utilize
second-generation technologies with attractive characteristics that
could be commercially available in the 1990s. Extensive programs,
funded by the Department of Energy and private groups, are now under way
to prove those technologies at the pilot and demonstration plant stages,
and to address the engineering and design problems that must be solved
before commercial development is possible.
The Hygas process has an advantage because a pilot plant based on
this process has been operating since 1971, while the H-Coal pilot plant
is only now being constructed. Many successful tests have been run on
the Hygas pilot plant, although several problems remain to be solved,
including introduction of the high-pressure coal slurry into the
reaction vessel, maintaining proper reaction conditions in the
three-stage gasifier, and corrosion of vessel materials. In addition,
the construction and operation of the large, high-pressure reactors
envisioned for a commercial plant have never been carried out.
The H-Coal process faces similar problems of high-pressure slurry
operation and materials corrosion. In addition, the lifetime of the
hydrogenation catalyst may be limited and, if so, catalyst regeneration
techniques must be developed. Also, a reliable process for separating
the liquid products from unreacted char and ash has yet to be
demonstrated.
IX-43
-------
Although the Hygas and H-Coal processes were chosen for analysis in
Chapter IV, the implementation of the systems does not depend on the
successful development of those particular processes. Other technology
choices could provide the needed coal conversion. The Lurgi gasifi-
cation process, for example, is a commercially available technology that
could provide pipeline gas for residences and fuel-cell power plants.
In fact, that technology has been chosen for use in commercial coal
gasification plants proposed by several pipeline companies.
The SRC II process, for which a pilot plant is operating in
Ft. Lewis, Washington, is designed to produce low sulfur fuel oil along
with a naphtha by-product. Additional hydrotreating of those products
could probably be used to produce suitable turbine fuel or reformable
fuel-cell fuel, respectively.
Thus, other coal conversion technologies would enable the imple-
mentation of Systems 1 through 5, although at higher costs, lower effi-
ciencies, and possibly greater environmental impact than indicated in
Chapters V, VI, and VII. Overall, coal gasification is more likely than
coal liquefaction to be implemented on a commercial scale in the time
frame considered in this study, primarily because of the more advanced
state of coal gasification technology, the widespread demand for the
product, and the high cost of alternative sources (e.g., imported LNG).
Thus, Systems 1, 2, and 5, which are based on coal gasification, are
favored over Systems 3 and 4.
The electricity generation technologies are the other key com-
ponents of the systems. Of the four types of technologies analyzed,
coal-fired power plants clearly have advantages because of their ease of
implementation, reliability, and operational experience. However,
compared to combined-cycle power plants and fuel-cell power plants, they
have slow startup and shutdown and are not very suitable for quick-
response, load-following applications. Generally, gas turbines used as
spinning reserves must provide that capability.
IX-44
-------
Gas turbine/steam turbine combined-cycle power plants are by now a
well-established component of the utility power generation base. Gas
turbines are among the more reliable electricity-generating devices, and
the heat recovery systems and steam turbines that constitute the re-
mainder of the power plant are highly reliable and use well known tech-
nology. Advances in combined-cycle power plant performance depend on
developments in gas turbine technology. To achieve the operating tem-
peratures and concommitant efficiencies discussed in Chapter IV will
require gas turbines that use ceramic vanes and blades in the expander
that can withstand the thermal shock associated with cycling at high
temperatures. Because gas turbines are continuously undergoing develop-
ment for aircraft and industrial applications, as well as for power
plant use, it seems likely that higher temperature operation can be
achieved.
The use of coal-derived liquids in gas turbines is under investi-
gation by the Department of Energy. Although no actual tests have been
carried out, some coal liquids may be suitable turbine fuels, although
additional processing may be required in some cases to increase the
hydrogen-to-carbon ratio and reduce viscosity.
Because fuel cells are a new technology in power plant appli-
cations, they are at a disadvantage compared to coal-fired power plants
and combined-cycle power plants. Once implemented, however, they offer
a number of operational advantages, as discussed in Chapter II (e.g.,
ease of load-following, low maintenance). If the demonstration of
first-generation fuel-cell power plants in utility applications proves
successful, and if stack lifetime goals are achieved, then the oper-
ational basis for implementing fuel cells will have been established.
There would remain, of course, the accomplishment of a successful mar-
keting and commercialization strategy, the consideration of which is
beyond the scope of this study.
IX-45
-------
If first-generation (phosphoric acid) fuel cells can be success-
fully marketed, then the advantages inherent in second-generation
(molten carbonate) cells should lead to even more widespread utili-
zation. As discussed in Chapter II, the main technological problems to
be overcome are the stability of the stack materials under cycling
conditions at high temperatures, seal corrosion, and electrode sinter-
ing. These problems are being addressed by ongoing programs sponsored
by the Department of Energy and private groups.
The use of fuel cells in on-site power generation applications, as
envisioned in System 5, is attractive in many respects, but has certain
operational disadvantages, including the need for load management to
avoid having to oversize the power plant to meet all conceivable loads,
and the requirement for reasonably good matching between electrical and
thermal loads. The implementation of such systems depends largely on
consumer aceptance of the concept, as well as on finding conditions
under which they will be economical.
A final key component in the energy systems is the advanced heat
pump described in Chapter IV. Although heat pumps have been com-
mercially available for many years, only recently have they achieved a
level of reliability that will make them widely acceptable in resi-
dential heating and cooling applications. With wider markets, the R&D
required to achieve the advances described in Chapter IV should be
readily justifiable to companies that manufacture heat pumps. The
technological requirements are relatively simple. The successful de-
velopment and marketing of advanced heat pumps will make electrically
based residential energy systems considerably more attractive compared
to gas-based systems than they are now.
In summary, System 1 has the fewest technological barriers to
overcome, is the most likely to be implemented, and can provide the
needed energy requirements in a satisfactory and reliable manner.
Systems 2 and 4 are approximately comparable in their difficulties of
implementation, which center on the fuel cell and coal liquefaction
IX-46
-------
components, respectively. Once implemented, System 2 will have an
advantage because of the operational benefits of the fuel-cell power
plant. System 5 must rank somewhat lower than System 2 because of the
operational disadvantages of the on-site power plant, as discussed
previously. Finally, System 3 appears to have the greatest relative
difficulty of implementation because it contains both coal liquefaction
and fuel-cell components. Assuming this system were implemented, its
operational advantages would be similar to those of System 2. However,
those advantages would probably not outweigh the difficulties of imple-
mentation.
From the preceding discussion, the ranking of the systems based on
their performance characteristics are as follows, from highest to
lowest: Systems 1, 2, 4, 5, and 3.
IX-47
-------
E. References—Chapter IX
1. National Academy of Sciences, "Particulate Polycyclic Organic
Matter," (Washington D.C., 1972).
2. Electrical World, various issues.
3. General Electric Company, "Transmission Line Reference Book — 345
kV and Above," Electric Power Research Institute.
IX-48
-------
X. SUMMARY AND CONCLUSIONS
The comparative analyses carried out in Chapter IX enable the
relative advantages and disadvantages of the five systems to be
determined. A tabulation of ordinal rankings of the five systems in
each category examined in Chapter IX is shown in Table X-l. In the
table, a ranking of 1 indicates the most desirable system (i.e., the
lowest cost, highest efficiency, lowest environmental impact), a ranking
of 2 the next most desirable, and so on. Using Table X-l as a guide,
the relative advantages and disadvantages of each system are summarized
below.
A. Summary of Advantages and Disadvantages
1. System 1 (Coal-Fired Power Plant; SNG)
System 1 has the lowest heating and cooling costs, as well as the
lowest capital cost of any of the five systems. In addition, this
system is most likely to meet the required performance standards,
provide residential energy reliably, and be widely adopted by utilities,
primarily because it employs the fewest number of new or advanced
technologies.
On the other hand, System 1 is the least energy-efficient of the
five systems, requiring 83% more energy resources to provide the same
amount of heating and cooling than the most efficient system. In
addition, it has the largest environmental impact of any system, ranking
lowest in four out of six categories, including the important categories
of air pollution and solid waste.
X-l
-------
Table X-l
SYSTEM RANKINGS IN VARIOUS CATEGORIES
System 1 System 2 System 3 System 4 System 5
Economics
Operating
Cost
Capital
Efficiency
Environment
Air
Water
Solid Waste
Noise
Land Use
Aesthetics
System
Performance
5
4
5
3*
5
5
1
2
2*
2*
2
3
2
2
3
2*
2*
5
2
3
4
4
5
2*
3*
4
4
3
1
1
1
1
1
1
5
Rankings are essentially equal. Differences are too small to be
resolved.
X-2
-------
2. System 2 (26 MW Fuel Cell - SNG)
Although System 2 ranks about midway in its cost of heating and
cooling, it requires almost the highest initial capital investment. Its
overall system performance is second only to System 1, and it is also
second highest in energy efficiency. It also ranks highly in nearly all
environmental categories except land use.
3. System 3 (26 MW Fuel Cell-Naphtha)
Because of its similarity to System 2, System 3 ranks closely with
that system, although its rankings are somewhat lower in most
categories. It is somewhat less costly to install than System 2, but
somewhat more costly to operate. The most noticeable difference between
the two systems is that System 3 ranks considerably lower in the noise
and system performance categories.
4. System 4 (Combined Cycle Power Plant)
System 4 has advantageous capital and operating costs compared with
most other systems, and ranks midway with respect to system performance
criteria and solid, waste generation. However, it ranks next to last in
terms of energy efficiency, and has low ranking in four of six
environmental categories.
5. System 5 (100-kW Fuel Cell with Heat Recovery)
The rankings for this system present the most interesting picture
of the five systems, because its rankings occur only in the extreme
categories, 1 or 5. It ranks highest in all noneconomic categories
except system performance, in which it ranks last. It also ranks last
in both capital and operating cost, but as shown in Chapter IX, the
operating costs are sensitive to the electrical and thermal loads that
the system is required to meet.
X-3
-------
B. Conclusions
Although we cannot categorically state which system is "best" or
"worst" overall, several important implications result from our system
analyses, especially in regard to the desirability of the implementation
of energy systems based on fuel cells. However, those implications are
applicable only to coal-based, intermediate-load electricity generation
for residential energy use as covered in this report and should not be
considered necessarily applicable to other types of systems such as
those based on different fuel sources or different end uses.
First, economics is not a driving force for implementing the three
fuel-cell systems (Systems 2, 3, and 5). Even under optimistic
assumptions about the cost of fuel cells, those systems are not
competitive with the alternatives. When only the cost of heating and
cooling is considered, System 5 could be competitive with the next least
costly system (System 4) under the appropriate conditions (high
thermal-to-electrical demand ratio), but the cost of supplying all
residential energy requirements is still very high compared to the other
systems, and substantial optimization procedures would have to be
carried out to determine the most economical applications.
Fuel-cell systems are more energy-efficient than the alternatives,
partly because of the high efficiency of fuel cells and their potential
for heat recovery, and partly because of reduced transmission losses
resulting from dispersed siting. Thus, energy resources — in this
case, coal — are conserved. Although coal is not as limited a
resource as petroleum and natural gas, its conservation is clearly
beneficial because it minimizes social and political pressures resulting
from increased coal mining in the West, and it extends the lifetime of
the most accessible, lowest cost coal reserves. High system
efficiencies could convey economic benefits as well, but only at coal
prices considerably higher than those currently in effect for western
surface mining. For example, System 5 would have a lower cost of
heating and cooling than System 1 only if the price of coal were at
X-4
-------
least ten times higher than that derived in Chapter VII ($0.33/GJ or
$6.69/tonne).
In addition, the three fuel-cell systems have less environmental
impact, primarily because of the low emission rates of the fuel-cell
power plants, which particularly benefit areas where pollutant loadings
already approach or exceed those allowable by law. Recent environmental
legislation, such as the Clean Air Act Amendments of 1977, clearly
favors the siting of power generation facilities with the lowest pos-
sible pollutant emission rates in nonattainment and "prevention of
significant deterioration" areas. In areas removed from power plant
operation, such as coal resources areas where conversion plants are
located, lower pollutant outputs per unit of energy ultimately consumed
will also be beneficial. When pollutant loadings begin to exceed statu-
tory limits in those areas, new conversion facilities will have to be
located elsewhere, entailing greater costs for coal shipment, construc-
tion of additional rail lines, and so on.
Furthermore, other environmental attributes, such as lower land-use
impact of fuel-cell systems, generally mean that the siting of power
plants and related facilities (e.g., transmission lines) is easier than
for alternative systems.
A full quantitative assessment of the environmental benefits of
fuel-cell systems was not possible in this study, and indeed, such
benefits are very site-specific. They depend heavily on such factors as
the local pollutant loadings at the time of implementation, local land-
use characteristics, and the availability of suitable sites for solid-
waste disposal.
Finally, the fuel-cell systems considered here have a range of
system performance characteristics depending on fuel type and appli-
cation. If fuel-cell demonstration programs and early commercial use
show that fuel cells will perform as projected in terms of load-
following, reliability, and stack lifetime, then implementation of
X-5
-------
fuel-cell systems will be greatly enhanced. However, new energy
technologies are often justifiably met with some distrust and
skepticism, especially by utilities, who must be concerned with the
reliability and performance of the electric power system and who may not
be willing to take even small risks when older, more familiar
technologies are available.
Ultimately, the utilization of fuel-cell power plants in various
energy systems will be driven not by economic considerations, but
primarily by environmental and operational ones. Those utilities
constrained by environmental, siting, and other noneconomic factors
should find fuel-cell systems attractive alternatives to other methods
of power generation.
As is the case for most promising, but yet unproven, energy
technologies, the support of the federal government will be important to
the ultimate success of fuel cells as a commercially viable concept.
DOE, EPA, NASA, and other federal agencies have provided considerable
funding for fuel-cell development. This financial support, which has
increased substantially in recent years, will help to ensure the
technical success of fuel-cell R&D programs. However, to ensure success
in the marketplace, additional steps will have to be taken relatively
early in the commercialization process, to assure that momentum is not
lost and that companies manufacturing fuel cells have a market for their
first, more costly units. As production increases, and costs decline,
conventional market forces should result in wider market penetration.
Steps that the government could take to assist in early
commercialization, subsequent to successful demonstration of first
generation fuel cells, include:
o Purchase of fuel-cell power plants for use in government
installations such as military bases.
o Additional investment tax credits and/or loan guarantees for
utilities that purchase fuel cells.
X-6
-------
o Incentives for the use of on-site fuel-cell power plants with
heat recovery in federally-funded housing projects.
o Legislative initiatives to ensure that innovative electrical
generation technologies such as fuel cells will be allowed to
use natural gas and petroleum fuels until the time when
synthetic fuels become available.
Prior to implementation of these actions, extensive cost/benefit
analyses should be undertaken to ensure that the benefits that accrue
from the implementation of fuel cells (fuel savings, environmental and
operational) are justified in terms of the public and private expen-
ditures required to achieve them.
In the coming years, all new power sources will be subject to
intense scrutiny by the government, by environmental groups, and by the
general public. They will be required to meet strict environmental
standards, yet provide electric power efficiently, reliably, and at an
acceptable cost. It appears that energy systems based on fuel cells
will be among the most likely to meet those requirements.
X-7
-------
Appendix A
ENTHALPIES BASED ON THE GIRDLER CATALYSTS DATA HANDBOOK
The enthalpies given in Sections IV-B and IV-C of this report are
based on information tabulated in Girdler Catalysts, Physical and
Thermodynamic Properties of Elements and Compounds, published by the
Girdler Company, Louisville, Kentucky. This book is a standard
reference for process and reactor engineering.
The Girdler data use a reference state of zero for the enthalpy of
the elements at a temperature of absolute zero. This standard state is
different from that found in many other references, such as Perry\s
Chemical Engineer's Handbook, but it is extremely useful. The
tabulations allow rapid enthalpy calculations and intermediate
temperatures are easily interpolated. Enthalpy balances on reactants
and products give the heat of reaction directly. Furthermore, because
having a process stream below the standard state is impossible, many
confusing sign changes are avoided. It is important, however, that all
enthalpies in a given calculation are based on the standard state.
1. Naphtha Enthalpy Calculations
The enthalpy of the coal-derived naphtha feed for System 3 was
converted to the Girdler Catalysts basis after assuming the following
enthalpic properties:
AH = Heat of Combustion = 46.8 MJ/kg (20,100 Btu/lb)
CNa
AH^ = Heat of Vaporization = 326 kJ/kg (140 Btu/lb)
Na
C = Heat Capacity (gas) = 1.65 kJ/kg-°c (0.395 Btu/lb-op)
Na @ 16°C (60°F)
A-l
-------
Note that the heat capacity was varied with temperature proportional to
the heat capacity of toluene, as given in Girdler Catalysts. The
enthalpy of the naphtha at the standard state is given by:
AH =2^ AIL - 5^ AH
Na 1 products 1 reactants
where n. is the number of moles of products or reactants, and AH. is
the enthalpy of products or reactants.
Based on lOOg of naphtha: g atom C = 7.11
g atom H = 14.60
g atom 0 = negligible
The reaction equation is thus:
C7.11 H14.60 + 14'41 °2 ~7'U C°2 + 7'3° V-
Below is a tabulation summarizing the naphtha enthalpy calculations:
Enthalpy n^ (g mole) AH£ (kJ/g mole) n£AH? (kJ)
AHH20
AH&>2
AHO
°2
AHvH2o
AHC
7.30
7.11
14.41
7.30
100.00 g
-229.0
-384.0
8.4
-442.0
-46.8 kJ/g
-1670
-2730
121
-3230
-4680
(liq., 16°C) = -1.77 kJ/g -177 kJ/lOOg
In the tabulation above, AH^ is the total enthalpy of the stream or
component i, and AHy is the heat of vaporization of component i.
A-2
-------
Enthalpies of naphtha as a vapor at 16°C were calculated as follows:
(gas, 16°C) = AH^a (liq. 16°C) +
vNa
= -1.77 + 0.33 = -1.44 kJ/g
Enthalpies of naphtha as a vapor at other temperatures were calculated as
follows:
AH^a (gas, 38°C) = AHJJ (gas, 16°C) + (38°C - 16°C) x Cp (16°)
Na
= -1.44 + (22) (0.00165) = -1.40 kJ/g
Enthalpies at other temperatures were calculated by numerical integration,
using appropriate heat capacities.
2. SR-52 Enthalpy Program
A simple matrix program was written on a programmable hand
calculator to facilitate the rapid calculation of stream enthalpies.
The program calculates enthalpies at intermediate temperatures by a
straight line interpolation. The program requires that stream flow
rates, Girdler Catalysts enthalpies and the temperature of the stream be
entered into the calculator before each calculation, because the storage
capacity of the SR-52 is limited. The basic equation of the program is:
o , n0 v o o
AH = (ni ) ( AH (T) . . . H A(T))
stream \nn / 1 n
where n. is expressed in g mole/hr, and H.(T) is expressed in
kJ/g mole.
A-3
-------
Appendix B
MOLTEN CARBONATE FUEL-CELL PERFORMANCE
A study of molten carbonate fuel-cell performance was undertaken to
determine what factors contribute to cell power output. This study
developed a calculational model that can reproduce published data on
molten carbonate fuel-cell performance.
1. Cell Performance
The survey of the literature ~ indicated that at moderate
current densities (i.e., well removed from diffusion limits), cell
performance can be approximated by the equilibrium cell voltage between
the anode and cathode streams minus a term proportional to the current
density. This term is mostly internal resistance of the cell, although
at low current densitites, varous electrode and electrolyte
polarizations also can appear to be linear with current density. A
study of this model shows that:
o Cell voltage increases slightly with increased total pressures.
o Cell voltage is decreased by the presence of diluents.
o As the gas moves along the anode or cathode, the reactants are
consumed and the gas is less reactive, so that at constant
cell voltage the current density decreases down the length of
the cell.
o Increasing the electrolyte thickness decreases cell voltage by
increasing the internal resistance.
o H2 is the most active anode reactant. However, CO is
continuously shifted by 1^0 to form more H2 so that a mole
of CO is virtually equivalent to a mole of H2 in the anode.
B-l
-------
o CH^ is inert in the anode at these temperatures.
o At the cathode, CC>2 has a more dramatic effect on the
voltage than oxygen because the voltage is proportional to
In (C02) and In (C^)^. Therefore, both excess 02 and
C02 are necessary for good cathode performance.
The features of the model described above were computerized on a
programmable hand calculator. This program can predict the voltage and
current density of a molten carbonate fuel cell with varying anode and
cathode feeds, pressures, fuel utilizations, and cell resistances.
After the local activity of an electrode was reduced to a single
resistance parameter, it was still necessary to do extensive calculation
to determine the performance of a total cell. The anode is required to
run at high conversion levels of the total CO and H in the feed gas,
so that the equilibrium potential varies considerably with distance down
the length of the cell. Therefore, the calculation procedure described
in greater detail in Section 2 of this appendix was developed. It
consists of the following steps:
(1) The feed gas was equilibriated with respect to the water gas
shift reaction: CO + 1^0 5=^ H2 + C02
(2) The equilibrium voltage relative to all components at unit
activity was calculated:
AE = (RT/2F)jhi[(C02)(H20)/(H2)J .
(3) The local current density was calculated from the cell
polarization minus the local polarization via
i = (Tj-AE)/ (Rint + Rext^- Rint was set at
0.95 ohm-cm^ for present technology and Rext at 0.3
ohm-cm^ for a total of 1.25 ohm-cm^.
(4) Using this current density, the distance required to convert a
quantity of hydrogen equal to 10% of the total H2 and CO
remaining was calculated.
(5) Also computed was the change in pressures and flow rates due
to the production of steam and C02 via the electrochemical
reaction: H2 + 003 •- H20 + C02 + 2e. The
program then recycled to Step (1).
B-2
-------
The calculation was repeated for 2, 6, 12, and 18 cycles to obtain
19, 47, 72, and 85% conversion of the gases. The average current
density was calculated from the cumulative current densities and
distances. The computations were repeated at various polarizations to
obtain performance curves.
These detailed calculations were not repeated at the cathode
because cathode gases are run to only 50% conversion. A good
correlation could be obtained with the data reported by Ackerman on
methane reformate at 50% air conversion, using the mean of the cathode
polarizations at the inlet and outlet of the cell. Because the cathode
[m
(C02).(02) , the mean
polarization is the log mean gas pressure combination. At very low
current densities and low anode conversions, the model breaks down
slightly, but at higher conversions, and current densities between 50
2
and 200 mA/cm , the results seem to be useful and reliable.
2. High Conversion Performance of Molten Carbonate Fuel Cell
The first step in the calculation is the determination of the
extent of the water-gas shift reaction:
CO + H20 5=i H2 + C02
or, abbreviated: A + B ^=* C + D.
If N is the loss in pressure of the CO or H20 due to the shift,
then
(C + N) (D + N)_
(A - N)(B - N) K, the equilibrium constant
K was taken at 1.85 in this work, although more careful
interpolation shows that at 650°C (1,202°F) a value of 1.92 would be
more appropriate:
N = Y - N/Y2 - 4(K-lXKAB-CD)
where Y = K(A+B) + C+D.
B-3
-------
N was then subtracted from the pressures for CO and HO and added
to those of H_ and CO^ to obtain the equilibriated pressures.
Next the equilibrium "polarization" of the anode was calculated.
RT
(C02)(H20)
(H2)
The local current density was calculated via
U = (T7 -TJ)/R
Here 77 is the constant applied polarization and R is the
combined internal and external resistance.
The distance increment, AX, was not fixed at the start but was
calculated so as to consume a charge AQ equal to 10% of the remaining
combustibles, H and CO.
V
g
AQ =
10PTOT
V is the gas flow rate expressed as mA/cm where all gas
O
molecules are taken as containing 2 electrons per mole.
The starting gas flow rate can be chosen arbitrarily because the
cell length will vary accordingly and the calculated current density
will not change. However, for convenience, the inlet gas flow rate was
taken as 100 . PTOT/[(CO) + (H2>]. Then AQ reads directly as
percent conversion.
AX = AQ/U
AX and AQ are summed cumulatively into storage registers to give
the total cell length and total conversion, respectively. The average
current density at any point is given by SAQ/AX.
B-4
-------
After each increment of hydrogen conversion the various gas
pressures must be adjusted because of the consumption of hydrogen and
the formation of steam and CC>2, and the increase in total gas flow
rate, all due to the reaction:
H2+ C°3~^H2° + C°2+ 2e
Thus, first (H2) is decremented by AQ . PTQT/V while (H20)
and (C02) are incremented by the same quantity. Then V is adjusted
• 9
via V = V + AQ and all gas pressures are reduced by the factor
• 5
v;-
Finally, all gas pressures are fed back into the water-gas shift
subroutine and the whole process is repeated.
The process is halted after 2, 6, 12, and 18 cycles to give 19, 47,
72j and 85% conversion. At each point the corresponding average current
densities and gas compositions may be obtained. The whole process may
be repeated with other values of TJ . The cell voltage corresponding
to a value of TJ is given by
E cell (mV) = 1015 + TJ - TJ
c o
Here, 1,015 mV is the theoretical cell potential at 650 C
(1,202°F) and 101 kPa (1 atm) pressure of all reactant gases.
TJ is the mean cathode potential calculated from the average of
c f kl
(RT/2F)£n (COJ.CCL) at the cathode inlet and outlet.
B-5
-------
3. References—Appendix B
1. J. P. Ackennan, "Molten Carbonate Fuel Cell Systems - Status and
Potential," Paper No. 391, National Electrochemical Society, 151st
Meeting, Philadelphia, PA, May 8-13, 1977.
2. J. M. King, "Energy Conversion Alternatives Study - United
Technologies Phase II Final Report," NASA CR-134955 (October 1976).
3. H. Selman, et al., Abstract No. 393, Electrochemical Society, 151st
National Meeting, Philadelphia, PA, May 8-13, 1977.
4. Institute of Gas Technology, "Fuel Cell Research on
Second-Generation Molten Carbonate Systems," Project 8984,
Quarterly Status Report (Jan. 1 - March 31, 1977).
5. H. A. Liebhafsky and E. C. Cairns, Fuel Cells and Fuel Batteries,
Chapter 12 (Wiley, New York, 1968).
B-6
-------
Appendix C
ENERGY SUPPLY/DEMAND PROGRAM FOR RESIDENCES SUPPLIED BY
THE 100-kW FUEL-CELL POWER PLANT
A Texas Instruments SR-59 calculator was programmed to calculate
hourly average energy parameters associated with the use of a 100-kW
fuel-cell power plant to supply electricity and heat to townhouse
residences. The basic features of that system were described in Chapter
IV. For each residence, the program calculates the hourly average
electrical load, heat pump duty factor, and recovered fuel-cell heat
delivered to the space heating system.
The inputs to the program are the hourly average external
temperature, heating load, hot water load, and light and appliance
electrical load. Fuel-cell power plant, heat pump, and heat exchanger
performance parameters are stored internally.
The operation of the program is illustrated by the flow chart shown
in Figure C-l. The variables shown in the flow chart are defined below
(all parameters represent hourly averages).
T: External temperature, °F or °C.
Q: Space heating load, Btu/hr or kJ/hr.
HW: Domestic hot water (DHW) load, Btu/hr
or kJ/hr.
E: Light and appliance electrical load,
kW.
E1: Total electrical load, kW.
HF(E or E1): Fuel-cell heat recovered as a
function of electrical load, Btu/hr
or kJ/hr.
C-l
-------
1. ENTER T, Q, HW, E
2. HF HFIE+0.21)
3. HD HD(HF.O)
4.
AQ
Q--f^(H
F - HW)
6.
DF 1.0
DF 0
E' E + 0.21
HDN = -^- (HF HW)
7.
El
HO
EKT)
HO(T)
8.
E = E + El
9.
HF HF(E )
1 1a
13a.
14a.
HDN DF
10.
: - HW)
HD HD(HF.HO)
HDO HD(HF.O)
12a. A Q - HDN - DFXHO
Yes
DF = (Q— HDNl/HO
E' E + 0.21 + DFIEI - 0.21)
HDN = -T^T- (HF - HW)
H r
12b. A Q-HDN—DFXHO
Yes
13b.
14b.
DF (Q— HDNl/HO
E' " E + El + (DF- DHO/3413
15. PRINT E , DF, HF, HDN
FIGURE C-1. PROGRAM FOR ENERGY SUPPLY/DEMAND CALCULATIONS
C-2
-------
EI(T): Heat pump electrical load as a
function of temperature, kW.
HO(T): Heat pump output as a function of
temperature, Btu/hr or kJ/hr.
HD(HF, HO): Heat delivered to space heating
system through heat exchanger as a
function of recovered fuel-cell
heat and heat pump output, Btu/hr
or kJ/hr.
HDN: Net recovered fuel-cell heat
delivered to space heating system,
after DHW demand is satisfied,
Btu/hr or kJ/hr.
DF: Heat pump duty factor — ratio of
net space heating demand to heat
pump output.
The operation of the program, as displayed in Figure C-l, begins
with the specification of the key input variables in Step 1. Steps 2-4
then determine whether recovered fuel-cell heat is sufficient to meet
both DHW and space heating loads without operation of the heat pump. If
so, the program ends and values of the output parameters are printed
(the value of E is increased by 0.21 kW, which is the fan power
requirement in the space heating system). If not, the remaining
variables are initialized in Steps 6-8.
The iterative part of the program begins in Step 9. The program
proceeds through either one of two independent branches, depending on
whether DF is greater or less than 1. Physically, the heat pump duty
factor can never exceed 1.0. However, for purposes of the program it is
allowed to do so, simply indicating that the net space heating load
exceeds the capacity of the heat pump and that electric resistance
heating must be added. The program ceases to iterate when the heat
supply (net recovered fuel-cell heat plus heat pump heat plus electric
resistance heat, if required) equals the heat demand, Q. This balance
C-3
-------
is considered to have been achieved when the difference between supply
and demand is less than 100 kJ/hr. If the heat supply and demand do not
balance, a new duty factor and electrical load (E1) are calculated and
the program iterates again.
C-4
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/7-79-105b
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Comparative Assessment of Residential Energy
Supply Systems That Use Fuel Cells (Technical
Report)
5. REPORT DATE
April 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHORS Rt V. Steele, D. C. Bomberger ,K. M. Clark,
R.F. Goldstein, R.L. Hays, M.
R. J. Bellows*. H. H. Horowitz,
*»• • t—'wvjhvy .»_, , ^^ ^ j-*^r**i PtSW*. g^W A ••*.*.« ATA * ^/ AM.J. *». •
R. F. Goldstein,R. L. Hays ,M. E. Gray ,G. Ciprios*
~ ' ~ " - - C.W.Snyder*
8. PERFORMING ORGANIZATION REPORT NO.
and
9. PERFORMING ORGANIZATION NAME AND ADDRESS
SRI International
333 Ravens wood Avenue
Menlo Park, California 94025
10. PROGRAM ELEMENT NO.
EHB534
11. CONTRACT/GRANT NO.
68-02-2180
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND P
Final; 9/76 - 1/79
D PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600/13
is.SUPPLEMENTARY NOTES ffiRL-RTP project officer is Gary L. Johnson, MD-63, 919/541-
2745. (*) Coauthors are Exxon personnel.
16. ABSTRACT
The report gives results of a comparison of residential energy supply sys-
tems using fuel cells. Twelve energy systems, able to provide residential heating
and cooling using technologies projected to be available toward the end of this cen-
tury, were designed conceptually. Only a few systems used fuel cells. All systems
used Western coal as the primary energy source, and all residences were assumed
to have identical heating and cooling demands typical of the mid-continent U.S.
After screening, five systems were analyzed in detail. The entire energy cycle,
from coal mine to end use, was examined for costs, efficiency, environmental im-
pact, and applicability. .The five energy systems are: (1) a coal-fired power plant
supplying electricity and a coal gasification plant supplying SNG; (2) a 26-MW fuel-
cell power plant fueled by coal-derived SNG supplying electricity; (3) a 26-MW fuel-
cell power plant fueled by coal-derived naphtha supplying electricity; (4) a combined-
cycle power plant fueled by coal-derived fuel oil supplying electricity; and (5) a
100-kW fuel-cell power plant fueled by coal-derived SNG, sited in a housing com-
plex, supplying electricity to heat pumps, with heat recovered from the fuel cell
supplying supplemental space heating and hot water. Results indicate that the fuel
cell systems are most costly, most efficient, and have least environmental impact.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
COSATI Field/Group
Pollution
Fuel Cells
Energy Conversion
Techniques
Residential Buildings
Heating
Cooline Systems
Assessments
Coal Gasification
Coal
Naphthas
Fuel Oil
Natural Gas
Heat Pumps
Pollution Control
Stationary Sources
Substitute Natural Gas
13B
10B
10A
13M
13A
14B
13H
21D
07C
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport/
Unclassified
21. NO. OF PAGES
482
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
D-l
------- |