EPA
Office of
Research and
Development
Industrial Environmental Research
Laboratory
Cincinnati, Ohio 45268
EPA-600/7-77-086
August 1977
PRELIMINARY ENVIRONMENTAL
ASSESSMENT OF SOLAR
ENERGY SYSTEMS
Interagency
Energy-Environment
Research and Development
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.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-77-086
August 1977
PRELIMINARY ENVIRDNME3SITAL ASSESSMENT
OF SOLAR ENERGY SYSTEMS
by
D. Richard Sears
Paul O. McCormick
Lockheed Missiles & Space Company, Inc.
Huntsville Research & Engineering Center
Huntsville, Alabama 35807
Contract No. 68-02-1331
Project Officer
Robert Hartley
Energy Systems Environmental Control Division
Industrial Environmental Research Laboratory
Cincinnati, Ohio 45268
INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
u.s. ELWIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Industrial Environmental Re-
search Laboratory, U.S. Environmental Protection Agency, and approved
for publication. Approval does not signify that the contents necessarily re
fleet the views and policies of the Environmental Protection Agency, nor
does mention of trade names or commercial products constitute endorse-
ment or recommendation for use.
11
-------
FOREWORD
When energy and material resources are extracted, processed, con-
verted, and used, the related pollutional impact on our environment and
even on our health often require that new and increasingly more efficient
pollution control methods be used. The Industrial Environmental Research
Laboratory Cincinnati (lERL-Ci) assists in developing and demonstrating
new and improved methodologies that will meet these needs both efficiently
and economically.
This report addresses the environmental consequences of three kinds
of solar energy utilization: photovoltaic, concentrator (steam electric) and
flat plate. The application of solar energy toward central power generating
stations is emphasized. Discussions of combined modes and of the geosyn-
chronous satellite generating stations are included. Numerous conclusions
and recommendations are developed. These should be useful to U.S. Environ-
mental Protection Agency (EPA) personnel concerned with environmental
quality related to power technology and conservation, to EPA and other
Federal agency personnel concerned with manufacturing processes in support
of solar energy development, and to Federal personnel assessing long-range
goals and tradeoffs consequent to other advanced energy developments.
For further information on these subjects, interested readers should
contact the Power Technology and Conservation Branch of the Energy Systems
Environmental Control Division.
David G. Stephan
Director
Industrial Environmental Research Laboratory
Cincinnati
111
-------
ABSTRACT
Central station solar-electric plants and flat plate space heating in-
stallations are environmentally superior to their respective conventional
alternatives because they produce little or no air and water pollution. Both
kinds of installations will require storage systems, also relatively clean
e nvi r onmentally.
Land area required for central station solar plants will be large, but
it is not as destructive or irreversible as with coal stripping. The eco-
logical impact of solar plants can be serious as a result of vegetation de-
struction. Visual effects can be extensive, with no mitigating technology.
Weather modifications may occur. Geosynchronous satellite generating
stations could be environmentally catastrophic from pollution caused by
large numbers of Space Shuttle launchers.
Some photovoltaic materials, such as gallium and cadmium may be
resource limited. Indirect effects, resulting from the production of large
quantities of photovoltaic materials, could be environmentally harmful.
IV
-------
CONTENTS
Foreword iii
Abstract iv
Figures „ vii
Tables „ ix
Acknowledgments xi
I Introduction 1
Scope and Objective 1
The National Plan 1
Socio-Economic and Political Considerations 2
International Aspects 2
Previous Assessments 3
II Conclusions and Recommendations 4
Conclusions 4
Recommendations 7
III Energy Projections and Goals 9
IV Photovoltaic Systems . . . . 16
Photovoltaic Power Generation — Technology
Orientation . . . . 16
Photovoltaic Receivers 16
Photovoltaic Storage Systems 19
Power Conditioning and Handling . . . 24
Central Station Photovoltaic Generation
on Geosynchronous Satellites 24
Combined Receivers at Load Centers 27
Concentrator Photovoltaic Systems 27
Environmental Assessment — Photovoltaic Systems ... 29
Siting and Grid Dispersal 29
Direct Effects 33
Geosynchronous Satellite Generating Station —
Experimental Problems 49
Hydrogen Cycle Storage — Direct Effects 54
Alternative Subsystems 54
Cryogenic Storage of Oxygen and H2 57
Hydrogen as a Fuel 57
Resource Commitment and Depletion 57
Indirect Effects 66
V Concentrator Systems 84
Concentrator Technology 84
Environmental Implications 89
Siting and Grid Dispersal 89
v
-------
VI Flat Plate Collectors (by P. CX McCormick) 93
Technogloy Orientation. 93
Current Technology 93
Effect of R&D Efforts on Flat Plate Collector Use . . 94
Solar Collector Manufacturers 95
Materials Used in Flat Plate Solar Collector
Systems 95
Storage Systems 96
Environmental Assessment 97
Siting 97
Direct Effects 97
Indirect Effects 100
Recommended Research 101
Summary 102
VII References 104
Appendixes
A Energy Forecast Background Information Ill
B Conversion Factors 119
VI
-------
FIGURES
Number Pag<
1 Comparison of electric generation with total domestic
energy consumption (Ref. 1) . 10
2 Conventional fuels consumption (Ref. 1) 11
3 Storage requirements and associated plant capacity
factor for capacity displacement. 20
4 Interfacing photovoltaic — fuel cycles (Ref. 23) 21
4
5 A conceptual 10 MW satellite photovoltaic power
station (Ref. 9) . . . .6. . . 25
6 Photovoltaic-thermal combined collector (Ref. 17)
(Reproduced by permission of author) 28
7 Combination photovoltaic and heating-cooling system
(Ref. 17). (Reproduced by permission of author) ...... 28
8 Isopleths of mean daily direct solar radiation, Langleys/
Day (Ref. 35) (1 Langley = 1 cal/cm2 = 3.62 Btu/ft2) ..... 29
9 Distribution of solar energy onto a horizontal surface
(figures give solar heat in Btu/ft2 per average day;
1000 Btu/ft3 = 276.24 Langleys) (Ref. 9) 31
10 Principle known regions of cavernous limestones
in the contiguous U. S 34
11 Groundwater areas in the U.S. (Ref. 39) 35
12 Comparison of land disturbed for 1 GWe coal-fired
steam electric plant and typical areas for 1 GW
solar plants (Ref. 9) 38
13 Thermal energy balance (Ref. 7) 43
14 Effect of roughness on boundary layer velocity (Ref. 23) ... 45
15 Unit operations detail for an electrolysis —fuel cell
hydrogen cycle (Ref. 27) 55
16 U.S. silicon production and demand projections, without
regard for major solar power applications (Ref. 62) 57
17 1968 supply-demand relationships for cadmium (Ref. 62)
(units in 1000 Ib or 454 kg) 59
VII
-------
Number Page
18 Comparison of U.S. trend projections and forecasts
for primary cadmium (Ref. 62) 60
19 U.S. gallium production and demand projections, without
regard for major solar power applications (Ref. 62) 61
20 U.S. arsenic demand and production trends, disregarding
solar development (Ref. 62) 62
21 U.S. lithium demand and production projections disregarding
solar and fusion power application (Ref. 62) 65
22 Soil contamination by airborne cadmium (Ref. 70) 73
23 Rates (tons/yr), routes, and reservoirs of cadmium
in the environment (Ref. 70) 77
24 Potential coal mining areas in the Western states (Ref. 77) . . 78
25 Solar thermal conversion: central receiver concept 85
26 Fixed mirror solar farm (Ref. 36) 86
27 Combined photovoltaic and thermal-electric system
(Ref. 23) 89
Vlll
-------
TABLES
Number Page
1 1985 Scenario Results (Ref. 1). 13
2 2000 Scenario Results (Ref. 1). . „ , . . . . 14
3 Energy Ranking of Technologies (Ref. 1) 15
4 National Ranking of R, D&D Technologies (Ref. 1) 15
5 Summary of Candidate Storage Modes 22
6 Projected Solar Power Station Deployment (Ref. 34) 26
7 Solar Radiation at Selected Locations in the United
States During 1970 30
8 Year 2000 Photovoltaic Power Capacity Projections
and Corresponding Land Commitment . 36
9 Typical Solid Fuel Composition for Space Shuttle
Launch Vehicle (Ref. 58) 51
10 Calculated Composition of the Inviscid Core of a
Typical SSLV Exhaust Plume 51
11 Significant Anticipated Direct Environmental Releases
from a 500 MW Electrolysis-Fuel Cell Hydrogen
Cycle Facility (Ref. 27) 56
12 Summary of Some Photovoltaic Materials Data 64
13 Energy Costs for Selected Products (Ref. 68) 66
14 Emissions (Ibs/MW hr) 68
15 Some Air Pollutants Produced Due to Energy Costs
of Materials Production for Solar Electric Facilities,
Per GW Installed Capacity 68
16 Estimates of Atmospheric Emissions of Cadmium
in the U. S. for 1968 (Ref. 70) 70
17 Composition of Ore Concentrate and Representative
Samples of Flue Dusts from Roasting and Sintering
Operations (Ref. 70) 71
18 Composition of Zinc and Cadmium Compounds in the Dust
from an Electrostatic Precipitator of a Roasting Plant as
Determined by Solvent Extraction (Ref. 70) 71
IX
-------
Number
19 Distribution of Cadmium in Product Streams from
Copper Smelters of Different Designs (Ref. 70) 72
20 Distribution of Cadmium in Bessemer Processing
of Matte from Reverbatory and Blast Furnace
Smelting (Ref. 70) 72
21 Hypothetical "Pollutants Prevented" by Solar Electric
Substitution for Coal Fired Utility Generation 75
22 Estimated Rates of Emission of Cadmium During
Production of Cadmium Products for 1968 (Ref. 70) 76
23 Fuel Mix at Electric Generating Plants in the
Western U.S. (Refs. 78, 79) 80
24 New Coal Mine Development in the West by 1985 (Ref. 80) . . 81
25 Workers in Photovoltaic Production: Project Independence
Accelerated Schedule (Ref. 23) . 81
26 Central Receiver Solar Power Plant Requirements
(Ref. 36) 88
27 Central Receiver Solar Plant Generation Potential
(Ref. 36) 88
28 Solar Heating and Cooling Materials Requirements 101
29 Flat Plate Collector Systems Summary 103
A-l Inputs for Scenarios (Ref. 1) 115
x
-------
ACKNOWLEDGMENTS
The authors are grateful for the cooperation and useful suggestions
and inputs furnished by members of the professional staffs of the U. S.
Environmental Protection Agency, Energy Research and Development Ad-
ministration, Lockheed, Electric Power Research Institute, and many in-
dustrial and academic organizations.
XI
-------
SECTION I
INTRODUCTION
SCOPE AND OBJECTIVES
This report is devoted to three solar energy utilization technologies:
photovoltaic, flat plate, and concentrator. Areas of application included are
residential, commercial, industrial and utility. Space and maritime appli-
cations are excluded.
Lockheed-Huntsville has prepared this report to assist the U.S. Environ-
mental Protection Agency (EPA) in: (1) identifying potential beneficial and ad-
verse environmental consequences of a major solar energy development; (2)
phasing its R&tD efforts so that timely and appropriate technical solutions and
regulatory postures may be developed in anticipation of major solar power de-
velopment, to help reduce reliance on ad hoc responses; and (3) identifying the
directions of current research efforts funded by other agencies, areas needing
additional EPA R&D effort and funding, and the developments likely to require
earliest concentration of effort in pollution abatement technology.
The third point, above, has been accomplished in part by compiling a
comprehensive list of Federally sponsored, solar-related research, which
will be submitted later as an appendix to this report. However, this report
stands by itself. A similar compilation of privately funded solar projects
might be useful to EPA but is not part of this task.
THE NATIONAL, PLAN
As part of a broadly based effort to develop our domestic energy re-
sources, various Federal agencies and their contractors have a mandate to
explore, develop, demonstrate, and encourage market penetration of a
variety of advanced energy sources. The lead agency is the Energy Research
and Development Administration (ERDA).
An overall strategy has been articulated by ERDA in two basic reports
constituting "A National Plan for Energy Research, Development and Demon-
stration," ERDA 48, Vols. 1 and 2 (Refs. 1 and 2). These reports constitute
the basic documents on the entire program, setting forth goals, strategy,
and implementation plans. Further elaboration of the Federal programs
specific to solar heating and cooling is found in ERDA-23A (Ref. 3). Other
approaches to solar energy utilization are presented in ERDA-49 (Ref. 4).
These four reports to the President and Congress were examined by the
Congressional Office of Technology Assessment (OTA), which issued its
analysis and critique in a fifth volume (Ref. 5). The latter volume includes
-------
some criticisms especially relevant to the concerns of EPA, In particular,
environmental tradeoffs and societal and institutional aspects of environ-
mental acceptability are discussed.
One must assume that these five documents fairly represent current
best guesses concerning goals, rates of development progress, demonstra-
tion, and market penetration for solar energy utilization as well as utiliza-
tion of other energy forms. To the degree that this assumption is correct,
they can be used by EPA to administer the phasing of its energy related re-
search and development.
SOCIO-ECONOMIC AND POLITICAL CONSIDERATIONS
Socio-economic analyses related to solar energy development and
marketing are extremely important and may indeed have significant impact
on environmental decision making. Examples are Federal tax incentives
for energy conservation in homes, small business administration encourage-
ment of solar-architectural enterprises, FHA appraisal rules relative to
conservation (e.g., storm windows), and local tax structures relative to
residence-sited solar heating and cooling installation. All these factors
influence the decisions of residential property owners and business investors.
A variety of other institutional, political, community motivation, economic,
and other non-technical matters more or less directly affect costs and in- '
centives, and thus market penetration.
It is recommended that these socio-economic and motivational factors
be studied by professionals in those fields and in community and regional
planning. We believe at least some of these studies should be conducted by
contractors to EPA and not in-house. These matters are excluded from this
report unless related very directly in a definable way to environmental
problems.
INTERNATIONAL ASPECTS
Matters relating to international relations are important and may have
environmental significance. For example, increased use of coal and uranium
instead of foreign oil, relations between foreign and domestic mineral re-
sources, possibilities of international collaboration in solar development,
and opportunities to export solar technology and high technology products —
all relate to goals-priorities, participation of non-Federal funds, etc. Of
these, only the matter of foreign and domestic mineral resources is dis-
cussed in this report, for some solar developments will severely strain or
exceed domestic resources.
Sunlight is not a "strategic material." Solar energy development seems
to have only a second-order effect on a nation's defense capabilities. Thus
a prime opportunity exists for truly international collaboration in solar energy
research. Likewise, most advanced nations are likely to have some concern
for the environmental consequences of advanced energy developments.
Unfortunately, little international collaboration occurs today (Ref. 6).
-------
PREVIOUS ASSESSMENTS
An enormous volume of recent literature exists in the area of solar
energy systems design, economics and utilization projections. Almost none
of the literature deals in any direct and substantative way with environmental
matters. There have been a few published environmental studies, however,
and more are currently in progress.
Dickson (Ref. 7) at the Stanford Research Institute and Griffith (Ref. 8)
at the EPA Industrial Environmental Research Laboratory, Research Tri-
angle Park, N. C., have both performed EPA-funded preliminary surveys.
The Atomic Energy Commission (now ERDA) included a cursory study
in the Environmental Statement for the Liquid Metal Fast Breeder Reactor
(Ref. 9).
Very recently MITRE Corporation (Ref. 10) completed three studies
sponsored by ERDA's Division of Solar Energy: a pilot solar thermal central
receiving station (not site specific); the Solar Thermal Test Facility at
Sandia; and a Solar Total Energy system at Ft. Hood, Texas. The project
officer at ERDA was J. W. Benson.
Lawrence Berkeley Laboratory (Ref. 11) has underway a $100K pro-
gram to study environmental aspects of solar and geothermal energy.
Approximately 40% of the effort is to be solar. Th±s is funded by ERDA.
Stanford Research Institute reportedly (Ref. 12) has a $195K ERDA-
funded program to examine long-term socio-economic effects of nine ad-
vanced energy technologies, including solar.
At EPA's Las Vegas research center, Donald Gilmore's (Ref. 13)
office is contemplating an environmental assessment project, to include
photovoltaic, flat plate, and power tower methods. Funding has not yet been
"identified."
In the private sector, the Electric Power Research Institute (Refs. 14
and 15), Palo Alto, Calif., has just comissioned two projects: Black & Veatch,
Kansas City, Mo., are under a 20-month, $323K contract to study five solar
energy technologies, including solar-thermal and photovoltaic. Also
Woodward-Clyde, Consultants, San Francisco, have undertaken a 19-month,
$293K project to study solar-thermal conversion and wind power using what
is described as an advanced assessment methodology to be developed by them.
Of these studies, only the original Dickson (SRI) study and the AEC-
LMFBR statement are readily available in the public domain. The others
are either internal agency documents, have yet to be initiated, completed
or issued, or are private and proprietary. As major solar projects progress
well into the planning stage, we can expect to see increasing numbers of
NEPA-mandated environmental impact statements published. The prelimi-
nary studies such as this report may suggest areas requiring detailed
attention in subsequent environmental impact statements.
-------
SECTION II
CONCLUSIONS AND RECOMMENDATIONS
CONCLUSIONS
1. Central station solar electric plants and flat plate space heating
installations are environmentally superior to their respective
conventional alternatives because they produce little or no air
pollution or water pollution (aside from the cooling tower effects
in solar steam/electric applications).
2. Little is known about the environmental impact of large-scale
manufacturing of photovoltaic materials and cells.
3. Three photovoltaic materials seem to show greatest near term
promise: Si, CdS and GaAs. Si will be used for the first pilot and
demonstration plants, probably with optical concentration. It is
quite unclear what photovoltaic materials and manufacturing
processes will prevail in the long term.
4. Solar generating stations will require storage in order to conform
to current utility industry operating philosophies. Pumped hydro
and batteries will see nearest term use. Favored battery concepts
now are Li-S and Na-S, but those are not sufficiently well developed
for immediate utility application. Consequently, large conventional
Pb-acid batteries may see near term use, together with pumped
hydro if siting permits.
5. Hydrogen cycle and flywheel storage systems are being developed
and ordered for small peaking applications. They may prove
satisfactory for large storage someday. They are clean environ-
mentally.
6. Solar photovoltaic (but not solar steam/electric) generation is par-
ticularly suitable for grid dispersed application. Significant energy
economies result from placing generating capacity near load centers.
This also may reduce currently perceived incentives for very high
voltage transmission lines.
7. Photovoltaic plants require no cooling water, no steam water, and no
process water, and therefore, cause no degradation of surface water
quality due to chemical additions or thermal plumes. This also makes
them especially suitable for water-short regions of the arid Southwest
8 Land surface alterations over very large areas (for either photo-
voltaic or steam/electric) could lead to perturbatxons in subsurface
hydrology.
-------
9. Estimates of land area required for central station solar plants
range from 4.5 to 30mi2/1000 MW. This land commitment is not
as destructive and not as irreversible as a commitment for coal
stripping. The 30 -/ear cumulative land area required for coal
stripping to support an equivalent capacity fossil plant ranges from
30% less to 500% more than that needed for solar plants.
10. Land prices will not soon be a determining factor in utility market
penetration by the solar industry, because currently photovoltaic
materials and optical and sun-tracking components are overwhelm-
ingly more important cost elements, and vast areas of low productivity
Federal lands are available in the U.S. Southwest.
11. Ecological impact of solar plants can be profound. The key element
in this impact is the alteration or destruction of vegetation directly
by construction or indirectly by shading.
12. Visual effects of solar plants can be extensive. Mitigating technology
does not exist .
13. Large-scale solar station developments could have some micro-
climate/weather modification impact through:
• Avoidance of particulate and COo emissions from fossil
fuel combustion
• Avoidance of cooling towers (photovoltaic only)
• Possible local effects on earth's albedo
• Reduced boundary layer velocities.
14. Photovoltaic installations (and to a lesser extent steam/electric)
seem to be less vulnerable to sabotage, terrorism, major fires,
and seismic events than are coal fired and nuclear installations.
15. Proposed siting of central station photovoltaic plants on geo-
synchronous satellites has some potential to become environ-
mentally catastrophic. Principal problems are:
o Very large number of launches annually with unconfined
combustion of huge quantities of propellant fuel. This
fuel will probably be NH.C^O. plus At. plus organic
nitrogen compounds.
• Beaming of power back to earth using microwaves
• Local noise effects, which could be avoided by resiting
the launch centers
• Possible use of nuclear -powered space tugs. In the event
of a Space Shuttle launch abort followed by disintegration
and burnup, nuclear materials could be released to the
atmosphere.
A major environmental benefit of the geosynchronous satellite sta
tion is the greatly reduced construction materials requirement.
-------
16. Silicon photovoltaic plants are not resource limited, but CdS and
GaAs probably are. CdS or GaAs needed for a single 1000 MW
photovoltaic represents a few percent of known U.S. domestic
reserves of Cd or Ga.
17. Glass, steel, aluminum, silicon, and lead energy costs are signifi
cant cost elements but not prohibitive.
18. The air and water pollution and soil contamination aspects of metal
smelting and refining have been studied intensively and continue to
receive regulatory attention. Shifts in product mix caused by major
solar demands for Ga, In, etc., may require new technology.
19. Environmental health and toxicity and perhaps some industrial
hygiene aspects of Ga, In and Sb are not well known. Practically
nothing has been reported on effects on aquatic biota.
20. Steam/electric concentrator systems share many advantages of
photovoltaic generation but differ as follows;
• Cooling towers and/or ponds are required.
• Being non-modular, grid dispersal is unlikely.
• Visual impact is greater-
• Weather modification effects may be more significant.
• No silicon, CdS, or GaAs are required.
21. Steam/electric installations are likely to become economically
justifiable much earlier than photovoltaic.
22. Liquid metals (e.g., Na) have been suggested for heat transfer
fluid and possibly thermal storage. Environmental consequences
of an accident involving large quantities of sodium are unknown.
23. Solar heating and heating/cooling technology is well understood.
Widespread application awaits major improvements in reliability
and cost pictures. Heat driven air-conditioning seems very un-
promising financially when not integrated into a heating/cooling
system.
24. Corrosion problems cause limited lifetime and acceptance of liquid-
in-aluminum or steel collectors. Capital costs have limited liquid-
in-copper.
25. Flat plate collection is not suitable for most industrial process
heat applications. It may find application to food and feed process-
ing and certain industrial applications needing low-grade heat.
26. Thermal storage in flat plate applications is absolutely essential.
Currently, water tank heat storage is used for liquid-in-metal
collectors, and rocks are used for air-in-metal systems. Phase
change materials (PCMs) and chemical change materials (CCMs)
are technically attractive but financially prohibitive in residential
application.
-------
27. Flat plate collectors probably cannot receive widespread use inside
large cities; suburban residential applications are more likely. Land
use commitment and air pollution potential are negligible or zero.
Glare can be offensive. Flat plate collectors may reduce the cur-
rently observed heat inbalance and heat bubble effects over major
metropolitan areas.
28. Groundwater and soil contamination by corrosion inhibitors (especi-
ally chromates), algaecides, and non-aqueous heat transfer fluids is
possible and could become severe if projected uses are correct
(3 x 10" Ib corrosion inhibitors and 4 x 10" Ib heat transfer fluids
by year 2000).
29. A major environmental advantage of all solar technologies is avoid-
ance of coal mining and fossil fuel combustion. Further, solar gen-
erating stations in the Southwest could relieve large energy shortfalls
caused by exhaustion of natural gas supplies. Flat plate collection
alone is projected to save 830 x 10" gal of oil in year 2000.
REG OMMEND ATI ONS
1. In collaboration with other Federal agencies, EPA should encourage
application of solar technologies.
2. EPA should perform studies of social, economic, and community
factors which will influence selections among various energy options,
especially solar, and of the socio-economic impact of major ad-
vanced energy developments.
3. EPA should perform environmental assessments of very large scale
manufacturing of following photovoltaic materials and cells: Si, CdS,
and GaAs. Similar assessments on other candidate materials should
be commenced as soon as it appears that they are likely to become
important.
4. Emphasis should be placed on effects of the new "exotics" on
aquatic biota. This refers especially to Ga, and secondarily
Sb and In.
5. EPA should perform environmental assessments of new energy
storage technologies — both central station and residential.
6. EPA should perform initial studies of possible microclimate/
weather modification effects of;
• Cooling towers in arid Southwest
• Boundary layer velocity changes
• Heat balance effects.
7. EPA should immediately begin a thorough review and assessment
of the environmental (especially air pollution) effects of Space
Shuttle launches supporting construction of satellite-sited photo-
voltaic generating stations.
8. EPA, ERDA, Interior, and DOD should review resource limitations
of photovoltaic and storage materials, especially Cd, Ga, Li, Pb, Cr.
-------
9. EPA and ERDA should review the recycle possibilities for these
same materials as well as As and In.
10. EPA should develop fine particle control technology for metal fumes,
especially Cd, Ga, GaAs, and also CdS.
11. EPA should perform intensive studies (independent of the utility
industry) of effects of air pollution on vegetation in the arid South-
west and high arid northern plateaus.
12. EPA should examine the environmental impact of the transportation
demands of large-scale central solar plant construction in the South-
west .
13. EPA should examine the consequences of major accidents involving
molten metal heat transfer media.
14. EPA should produce an approved list of residential and commercial
storage materials (including PCMs and CCMs) and aqueous corrosion
inhibitors.
-------
SECTION III
ENERGY PROJECTIONS AND GOALS
Reports ERDA 23A, 48 and 49 contain extensive ERDA estimates and
flow diagrams of energy production modes and consumption, and tabulations
of resource consumption, and pollutant release. These parameters are pre-
sented as functions of several "scenarios" in Figures 1 and 2 and in Tables
1 and 2. The background and descriptive information for these scenarios
are presented in Appendix A, projected to the year 2000. In none of these
scenarios is solar electric energy production conceived to reach 2% of coal
electric.
Figures 1 and 2 present overviews of the consequences of the various
scenarios. Particularly striking is the rapid approach to uranium exhaustion
in scenarios O, II and III even though III includes breeder reactor participa-
tion. Scenario I is most effective from nearly all points of view, and its
success is primarily due to improved efficiencies in end use, with only token
contributions from advanced energy sources. These scenario presentations
serve to illuminate our very restricted options. They suggest that an aggres-
sive conservation campaign should accompany any advanced energy develop-
ment program.
Tables 1 and 2 present quantitative resource commitment data, and
direct and indirect environmental residuals for 1985 and 2000. These tables
reemphasize that conservation is the most benign scenario, from the environ-
mentalist's point of view.
ERDA fails to specify the emission factors and other parameters used
to calculate the pollutant data. Further, the tables contain no entries for
heavy metals and other "exotic" pollutants which are potential problems of
some advanced energy systems. A goal of this report is to call attention to
potential problems with "exotic" residuals arising from solar energy exploi-
tation, particularly photovoltaic.
Finally, Tables 3 and 4 display the goals and priorities ERDA had
established as of June 1975. Even though solar electric can make only a
small contribution by the year 2000, our options are so restricted then and
later that ERDA has ranked solar electric among its highest priorities.
ERDA 76-1 is now scheduled for release in March 1976. It is possible that
newly revised estimates will appear in that document.
-------
CO
Q
ID 160 -
O
z
o 140
lo 120
O
o
>100
cc
80
0, II
Total Energy Consumption
SCENARIOS
No New Initiatives
Improved Efficiencies in End Use
Synthetics from Coal and Shale
Intensive Electrification
Limited Nuclear Power
Combination of All Technologies
1975 1980 1985 1990
CALENDAR YEAR
1995 2000
CO
Q
O 25
Z
•z.
9 20
cc.
01
15
5 10
QC
O
3 5
UJ
_L
Total Electric Generation
SCENARIOS
No New Initiatives
Improved Efficiencies in End Use
Synthetics from Coal and Shale
Intensive Electrification
Limited Nuclear Power
Combination of All Technologies
1975 1980 1985 1990 1995
CALENDAR YEAR
2000
Figure 1. Comparison of electric generation with total domestic
energy consumption (Ref. 1).
10
-------
50,
V)
1 40|
O
O 30
z
O
CJ
20|
10
INCLUDES METALLURGICAL
COAL EXPORTS
IV
1975 1980 1985 1990
CALENDAR YEAR
1995 2000
Coal Consumption
SCENARIOS
No New Initiatives
Improved Efficiencies in End Use
Synthetics from Coal and Shale
Intensive Electrification
Limited Nuclear Power
Combin.ition of A|| Technologies
-10
SCENARIOS
No New Initiatives
Improved Efficiencies in End Use
Synthetics from Coal and Shale
Intensive Electrification
Limited Nuclear Power
Combination of All Technologies
Imports of Oil and Gas
1975 1980 1985 1990
CALENDAR YEAR
1995 2000
Figure 2. Conventional fuels consumption (Ref. 1).
(Continued)
11
-------
3.6
U. S. URANIUM RESOURCES
CO
^
CO
H
O
3.0
2.5
cr 2.0
O
I
CO
u.
O 1.5
10
O
i! 1.0
.5
1975
0
I
II
III
IV
V
SCENARIOS
No New Initiatives
Improved Efficiencies in End Use
Synthetics from Coal and Shale
Intensive Electrification
Limited Nuclear Power
Combination of All Technologies
Quantity of Committed
Uranium
ASSUMING PLUTONIUM RECYCLE
AND 0.20 TAILS ASSAYS
I
I
I
1980
1985 1990
CALENDAR YEAR
1995
2000
Figure 2. Conventional fuels consumption (Ref. 1).
(Concluded)
12
-------
TABLE 1. 1985 SCENARIO RESULTS (REF. 1)
Hydroelectric (at 34% efficiency)
Geothermal
Solar
Fusion
Light Water Reactor (LWR)
Liquid Metal Fast Breeder (LMFBR)
High temperature Gas Reactor (HTGR)
Oil Steam F.lectric
Gas Steam Electric
Oil, Domestic and Imports
Oil Imports
Oil Shale
Natural Gas, Domestic and Imports
Coal (including 1.5 Quads exports)
Coal (million tons per year)
Waste Materials
Biomass
Total Energy Resources (including exports)
Total Cost in Billions of Dollars per year
Average Cost in Dollars per
Million Btu of Resources Used
Resources Consumed. Quads (10"Btu)
I II III IV
Year 1985 Scenario Results— Resources
3.38
0.69
0.00
0.00
10.61
0.00
0.24
3.39
4.39
47.14
25.94
0.00
24.00
21.14
1006
0.10
0.00
107.30
226.S3
3.38
0.93
0.25
0.00
10.61
0.00
0.25
2.79
3.00
34.59
10.49
0.00
26.50
18.46
879
2.00
0.00
96.97
198.17
3.38
0.69
0.00
0.00
10.61
0.00
0.24
3.39
4.39
41.43
17.33
1.00
26.50
23.28
1108
0.10
0.05
107.28
224.S4
3.38
1.60
0.31
0.00
12.97
0.00
0.24
4.91
3.19
41.57
17.47
0.00
26.50
20.10
957
0.10
0.00
106.77
223.74
3.38
3.20
0.57
0.00
10.60
0.00
0.25
2.32
4.03
41.52
17.42
1.00
26.50
19.98
951
0.00
0.05
107.05
218.57
3.38
l.SO
0.31
0.00
12.97
0.00
0.25
2.79
3.00
31.95
7.85
1.00
26.50
18.13
863
2.00
0.05
98.14
197.15
2.11
2.05
2.10
2.10
2.05
2.01
Year 1985 Scenario Results-Environmental Effects
Centralized Air Pollutants
Carbon Dioxide (CO,) 10" pounds
Carbon Monoxide (CO) 10" pounds
Nitrogen Oxides (NO,) 10° pounds
Sulfur Dioxide (S0;) 1O' pounds
Particulates 10* pounds
Hydrocarbons (HC) 10* pounds
Decentrali7fid Air Pollutants
CO, 10" pounds
CO 10: pounds
NO. 10° pounds
SO. 10' pounds
Particulates 10" pounds
HC Iff pounds
Total Air Pollutants
CO., 10" pounds
CO . J: pounds
NO. 10* pounds
SO, 10* pounds
Particulates 10s pounds
HC 10" pounds
Water Pollutants (all in 1000 tons)
Bases
Nitrates
Other Dissolved Solids
Suspended Solids
Nondegradable Organics
Biological Oxygen Demand (BOD)
Aldehydes
Radioactive Effluents
Solids. 1000 ft1
Krypton-85, 10" curies
Tritium, 10s curies
Population Exposure, 1000 man-rem
Heat Dissipated
Central Sources (Quads;
Decentralized "
Total
Solid waste, million tons
Land use, million acres
Occupational Health & Safety
Deaths
Injuries (1000s)
Man-Days Lost (1000s)
40.5
55.1
12.2
18.7
68.5
3.6
97.5
6339.6
24.8
16.0
160.7
133.8
138.0
6394.7
37.0
34.7
229.2
137.4
3.9
1.8
552.8
98.2
26.6
69.1
192.1
13.1
36.1
22.2
64.3
36-b
59.9
106.4
2569.7
17.2
209.0
11.2
540.6
36.2
52.0
11.1
17.4
64.7
2.9
80.7
5818.9
21.5
9.6
134.2
119.8
116.9
5870.9
32.6
27.0
198.9
122.7
3.4
1.8
481.7
86.2
19.5
57.5
146.0
13.1
36.1
22.2
64.3
33.6
Sl.l
94.7
1961.9
15.4
180.0
9.6
461.0
40.5
55.1
12.2
18.7
68.5
3.6
95.7
6223.1
24.2
13.2
157.1
132.2
136.2
6278.2
36.4
31.9
225.6
135.8
3.9
1.8
533.1
94.2
23.4
65.1
170.2
13.1
36.1
22.2
64.3
36.5
C9.9
106.4
2368.0
17.1
223.0
11.8
567.7
41.5
55.6
12.5
19.8
69.2
3.4
86.7
6129.5
22.6
11.3
137.3
129.8
120.2
6185.1
35.1
31.1
206.5
133.2
3.3
2.2
521.7
939
23.3
65.1
170.9
15.9
44.0
27.1
78.2
39.9
i*~>.^
105.4
2304.9
17.5
199.0
10.7
511.6
35.7
E0.5
10.8
16.5
62.3
3.2
91.4
6321.2
23.6
12.5
119.5
133.3
127.1
6371.7
34.4
29.0
181.8
136.5
2.6
1.8
471.9
88.3
23.5
62.3
168.8
13.1
36.1
22.2
64.3
06. i
03 .3
105.4
2297.2
16.7
196.0
10.5
504.7
30.6
42.C
9.3
14.3
52.2
2.6
80.6
5573.7
21.5
9.7
133.9
117.1
111.2
5615.7
30.8
24.0
185.1
119.7
3.4
2.2
438.1
71.8
17.7
55.3
131.1
15.9
44.0
27.1
78.2
34.1
02. u
96.1
1831.1
15.3
176.0
9.3
446.9
13
-------
TABLE 2. 2000 SCENARIO RESULTS (REF. 1]
Resources Consumed, Quads (l(V~Btij)
Hydroelectric (at 34% efficiency)
Geothermal
Solar
Light Water Reactor (LWR)
Liquid Metal Fast Breeder (LMFBR)
High Temperature Gas Reactor (HTGR)
Oil Stea.-n Electric
Gas Steam Electric
Oil, Domestic and Imports
Oil Imports
Oil Shale
Natural Gas. Domestic and imports
Coal (including 1.5 Quads exports)
Coal (million tons per year)
Waste Materials
Biomass
Total Energy nesources (including exports)
Total Cost in Billions of Dollars per year
Average Cost in Dollars per
Million Btu of Resources Used
0
3.65
1.40
0.00
0.00
36.59
C.OO
3.90
4.07
2.00
70.54
58.34
0.00
15.40
33.89
1614
0.10
0.00
165.47
493.94
3.02
1
Y»»r 2000
3.65
240
3.50
0.00
le.bo
0.00
3.&0
2.:c
0.00
40.32
20.62
0.00
22.30
22.91
1091
6.50
0.00
122.48
325.64
2.74
II
Scenario
3.65
1.40
0.00
0.00
36.S9
0.00
3.90
3.77
2.00
37. 71
18.01
8.00
22.80
49.77
2370
0.10
1.50
165.42
460.52
2.78
III
IV
V
Retuftt— Resources
3.65
6.60
6.59
0.05
36.59
3.90
3.50
4.08
2.00
46.47
26.77
0.00
22.80
30.51
1453
0.10
0.00
161.16
469.54
2.98
3.65
14.93
9.52
0.05
1097
O.OC
0.40
2.4-1
2.00
46.30
20.55
3.00
22.80
45.87
2184
O.CO
1.50
158.01
396.96
2.57
3.65
6.60
4.82
0.05
16.50
3.90
3.90
i.SS
0.00
1!).77
(4.11)
8.00
22.80
39.11
1862
6.50
1 50
137.03
328.74
2.46
Centralized Air Pollutants
Carbon Dioxide (CO,) 10" pounds
Carbon Monoxide (CO) 10' pounds
Nitrogen Oxides (NO,) 10° pounds
Sulfur Dioxide (SOj) 10* pounds
Participates 10' pounds
Hydrocarbons (HC) 10' pounds
Decentralized Air Pollutants
COi 10" pounds
CO 10' pounds
NO, 10" pounds
SOi 10" pounds
PartK'ulates 10' pounds
HC 10" pounds
Total Air Pollutants
CO, 10" pounds
CO 10' pounds
NO. 10» pounds
SO, 10* pounds
Particulates 10* pounds
HC 10" pounds
Water Pollutants (all In 1000 tons)
Bases
Nitrates
Other Dissolved Solids
Suspended Solids
Nondegradable Organics
Biological Oxygen Demand (BOD)
Aldehydes
Radioactive Effluents
Solids, 1000 ff
Krypton—85, 10* curies
Tritium, 1C? curies
Population Exposule, 1000 man-rem
Heat Dissipated
Central Sources (Quads)
Decentralized "
Total
Solid Waste, million tons
Land Use, million acres
Occupational Health & Safety
Deaths
Injuries (1000s)
Man-Days Lost (1000s)
Year 200O Scenario Results—Environmental Effects
42.5
62.9
13.0
20.9
76.6
3.1
136.4
9651.3
41.8
30.7
301.5
201.7
178.9
9714.2
54.8
51.6
378.1
204.8
9.2
6.9
1005.5
139.7
39.8
94.6
281.0
51.5
135.2
82.2
250.1
71.9
93.3
165.2
4001.8
27.5
361 0
18.7
920.6
26.0
43.0
8.2
13.6
51.3
1.5
100.7
7677.7
32.4
16.0
236.8
163.7
126.7
7720.7
40.6
29.6
288.1
165.2
7.4
3.5
705.2
100.2
22.8
58.2
159.6
26.8
65.9
39.5
128.0
43.6
71.3
114.9
2392.6
18.0
239.0
12.4
608.1
41.0
61.2
12.6
20.1
74.3
3.1
134.4
9608.5
37.7
17.5
202.5
171.8
175.4
9669.7
50.3
37.6
276.8
174.9
6.5
7.0
796.3
113.2
21.3
71.9
154.4
52.3
137.5
83.6
250.1
71.9
93.2
165.1
2972.4
26.7
480.0
23.5
1169.8
43.5
65.2
13.3
21.4
78.8
3.2
86.8
8875.6
33.1
13.6
181.2
185.1
130.3
8940.8
46.4
35.0
360.0
188.3
5.8
7.6
889.1
139.6
26.3
78.0
188.3
57.1
142.9
97.8
277.9
85.3
68.2
153.5
2887.7
28.9
322.0
16.5
808.0
34.9
53.2
10.8
17.1
65.1
2.6
138.4
9603.8
38.3
20.0
179.3
173.1
173.3
9657.0
49.1
37.1
244.4
175.7
5.2
1.9
611.8
104.7
22.7
65.8
162.6
14.3
39.1
24.0
69.3
50.0
97.0
147.0
2974.0
21.7
435.0
21.5
1072.6
23.7
38.9
7.5
12.4
46.9
1.4
106.7
E438.7
13.1
13.3
203.8
75.2
130.4
5477.6
20.6
25.7
250.7
76.6
6.7
4.1
525.3
58.2
8.9
41.1
64.0
31.6
71.2
53.7
155.8
45.9
77.3
123.8
1702.4
18.0
364.0
17.5
875.2
14
-------
TABLE 3. ENERGY RANKING OF TECHNOLOGIES (REF. 1)
TECHNOLOGY
TERM OF0
IMPACT
DIRECT"
SUBSTITUTION
FOR OIL
&GAS
R.DiD
STATUS
IMPACT IN"*
YEAR 2000
IN QUADS
GOAL I: Exoanded the Domestic Supply of
Economically Recoverable Energy Producing
Raw Materials
Oil and Gas—Enhanced Recovery Near Yes
Oil Shale Mid Yes
Geothcrmal Mid No
GOAL II: Increase the Use of Essentially
Inexhaustiole Domestic Energy Resources
Solar Electric Long No
Breeder Reactors Long No
Fusion Long No
GOAL III: Efficiently Transform Fuel Resources
into More Desirable Forms
Coal—Direct Utilization Utility/Industry Near Yes
Waste Materials to Energy Near Yes
Gaseous & Liquid Fuels from Coal Mid Yes
Fuels from Bicrnass Long Yes
Pilot
Study/Pilot
Lab/Pilot
Lab
Lab/Pilot
Lab
Pilot/Demo
Comm
Pilot/Demo
Lab
13.6
7.3
3.1 5.6
2.3-4.2
3.1
24.5
4.9
14.0
1.4
GOAL IV: Increase the Efficiency and Reliability
of the Processes Used in the Energy
Conversion and Delivery Systems
Nuclear Converter Reactors
Electric Conversion Efficiency
Energy Storage
Electric Power Transmission and Distribution
GOAL V: Transform Consumption Patterns to
Improve Energy Utilization
Solar Keat & Cooling
Waste Heat Utilization
Electric Transport
Hydrogen in Energy Systems
GOAL VI: Increase FnH-Use Ffficiency
Transportation Efficiency
Industrial Energy Efficiency
Conservation in Buildings and Consumer Products
Near
Mid
Mid
Long
Mid
Mid
Long
Long
Near
Near
Near
No
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Demo/Comm
Lab
Lab
Lab
Pilot
Study/Demo
Study/Lab
Study
Study/Lab
Study/Comm
Study/Comm
28.0
2.6
—
1.4
5.9
4.9
1.3
~
9.0
8.0
7.1
•Near—now througn 1985
Mid—1985 through 2000
Long—Post-2000
••Assumes no change in end-use device.
•••Maximum impact of this technology in any scenario measured in terms of additional oil which would have to be
marketed if the technology were not implemented. Basis for calculation explained in Appendix B.
TABLE 4. NATIONAL RANKING OF R, D&D TECHNOLOGIES (REF. 1)
Near-Term Major Energy Systems
New Sources of Liquids and Gases for the Mid-Term
"Inexhaustible" Sources for Hie Long-Term
Near-Term Efficiency (Conservation) Technologies
Under-Used Mid-Term Technologies
Technologies Support'ng Intensive Electrification
Technologies Being Explored for the Long-Term
Coal—Direct Utilization in Utility/Industry
Nuclear—Converter Reactors
Oil and Gas—Enhanced Recovery
Gaseous & Liquid Fuels from Coal
Oil Shale
Breeder Reactors
Fusion
Solar Electric
Conservation in Buildings & Consumer Products
Industrial Energy Efficiency
Transportation Efficiency
Waste Materials to Energy
Geothermal
Solar Heating and Cooling
Waste Heat Utilization
Electric Conversion Efficiency
Electric Power Transmission and Distribution
Electric Transport
Energy Storage
Fuels from Biomass
Hydrogen in Energy Systems
Highest
>. Priority
Supply
Highest
> Priority
Demand
Other
Important
Technologies
15
-------
SECTION IV
PHOTOVOLTAIC SYSTEMS
PHOTOVOLTAIC POWER GENERATION- TECHNOLOGY ORIENTATION
This technology has been applied extensively in small devices for re-
mote applications such as space, buoys, automatic control of street lamps,
microwave relay switching, etc. Nearly all of these applications have
features in common with the present application (widespread power genera-
tion) but none of them has the enormous terrestrial (land use) impact.
Further, the capital cost of the solar cells themselves has seldom approached
the overriding importance that it will in central electric generating station
application. These two related considerations— capital cost and size — force
cell efficiency to become an extraordinarily important design parameter.
A photovoltaic central electric generating station will have as its
principal components:
• The photovoltaic receivers ("panels," "cells")
• A storage facility
• Power conditioning equipment
• Transmission and switching equipment.
In contrast to conventional steam electric stations, water requirements are
trivial and the absence of reject heat eliminates the need for cooling towers.
Photovoltaic Receivers
Photovoltaic solar cells basically are semiconductor devices in which
photons are absorbed and charge carriers generated, together with a potential
barrier (such as a p-n junction) to separate the charges. These semiconduc-
tors must have several very specific properties for photovoltaic application.
An especially critical property is the energy gap, a function of both the
elemental composition and the crystal structure. Crystallite size is im-
portant. Some materials must be single crystals or annealed coarse grained
polycrystals, rather than fine polycrystalline deposits.
Typical photovoltaic materials are elemental or binary inorganic com-
pounds. A few higher order inorganic compounds show promise, and a few
organics have been suggested. Very high purity (often with "doping" with a
trace element additive) is required.
16
-------
Materials which satisfy all requirements have seldom been produced
in kiloton quantities. Therefore we often lack information on pollution con-
trol technology, ecological effects, toxicology and soil chemistry, for
example. Current new source performance standards, effluent guidelines,
and OSHA regulations may not be appropriate for all candidate compositions.
Standard methods of analysis may not be established for ambient air quality,
water quality and industrial hygiene situations. (This is now the case for Ga
and In.)
At the moment, industry research is moving vigorously in many direc-
tions; it is quite unclear what materials will prevail in the immediate term.
One expert (Ref. 16) believes single crystal silicon is the most promising,
feeling that 18 to 20% efficiency will be practical soon. An ERDA Photovoltaic
Branch spokesman (Ref. 6) feels that in the near term emphasis will be on
development of thin silicon (100 jUm, rather than current 250 /im) devices,
but that the intermediate term will see applications of thin polycrystalline
films using concentrators. He does not discount the possibility of ternary
materials being developed successfully. In the longer term, the same official
foresees possible abandonment of concentrators because of prohibitive costs
in very large scale application. An Electric Power Research Institute spokes-
man is certain that polycry stalline CdS will prevail (Ref. 14), but at a maxi-
mum efficiency of about 8% (Ref. 16). Some experts believe it will never be
cost competitive with silicon (Ref. 16). InSb, Se and group III phosphides
have adherents.
For reviews of current technology see Wolf (Ref. 17), Hickok (Ref. 18)
and the AIAA monographs (Ref. 19). Wolf reports active current R&tD di-
rections in the industry to be:
1. Deposition onto float glass
2. Development of GaAs cells with (Ga, A^)As window layers
3. Improved high temperature performance
4. Cost reduction by efforts in:
• Replacement of vacuum deposition by spray deposition
• Using lower cost raw materials
• Reduction in layer thickness
• Ribbon production of Si cells to replace discrete crystal
growth, sawing and lapping
• Production of Si ribbons by dendritic growth
• Chemical vapor deposition of silicon from volatile Si
compounds
• Vapor deposition of Si by crucibleless electron
beam heating
• Development of low cost substrates.
5. Development of concentrators to reduce cell area requirements
6. Development of automated production methods.
17
-------
The most cursory review of the literature reveals the overriding con-
cern with cost reduction. It is not clear why tradeoffs between cost per unit
area and efficiency are not pursued, but most sources resist such com-
promises. At this time, $1000/kW (1976 dollars) is considered a tolerable
cost. By year 2000, $1500/kW (1976 dollars) will probably seem attractive
(Ref. 16). What is clear is that the ultimately successful materials, and
production engineering technology, are almost wholly unsettled for the inter-
mediate and long term.
While it may be premature to institute extensive research on control
technology for production of all candidate materials, EPA should maintain
close surveillance on industrial practice and close liaison with ERDA's
Photovoltaic Branch so that it can anticipate significant trends in the industry
with ample lead time. A regulatory strategy probably can be developed with-
out reference to specific elemental constituents.
For the near term, we do know that the following compositions will be
used in pilot and demonstration installations:
• Si
• CdS
• CdS-CuS
• GaAs
• (Ga, A^)As
Requests for proposals and contracts are currently being issued (Ref. 6).
Currently, the Jet Propulsion Laboratory, Pasadena, Calif., is coordi-
nating 27 contracts for solar array manufacture.,
Research on CdS cells includes work on spray deposition, accelerated
life testing, hermetic sealing requirements and, as always, cost reduction.
CdS cells are degraded by atmospheric moisture and oxygen, and by solar
warming. The French pioneered improved lifetime of CdS cells by hermetic
sealing (Ref. 17).
Currently, there is one U.S. manufacturer of complete CdS panels
(Ref. 6): — Solar Energy Systems, Newark, Del. Their product is hermeti-
cally sealed and guaranteed for only one year.
It is important to distinguish between the various CdS products.
Pigment grade CdS is unsuitable for semiconductor application; its production
methods (Ref. 20) may conceivably be adapted to produce semiconductor
material, however, Photovoltaic and photosensitive CdS are two different
products (Ref. 21). A still more refined research grade (50 ppm) is supplied
by Eagle-Picher (Ref. 21) but is used only for laboratory scale studies.
Suppliers of CdS material include Fisher Scientific, Atomurgic, General
Electric, Sylvania, and Baker and Adamson. CdS devices are supplied by
National Semiconductor, Vactac, Clairex, General Electric, and Varian.
18
-------
Should CdS solar cell panels become economically attractive high demand
components, some of these material and device manufacturers may enter
the market, possibly with copper, float glass or Mylar™^ substrate.
Production and resource data for cadmium products are discussed as
environmental effects in a later section.
Varian is a major supplier of GaAs. Manufacture of the compound is
currently a very low volume business. Production and resource data are
discussed later.
Silicon solar cell manufacturers include Heliotech, Centralab, Solar
Power Corporation (Exxon), Sharp (Ref. 22) and Spectrolab (Textron) (Ref.
23). A description of discrete silicon solar cell manufacture is provided by
Spectrolab in the Project Independence report (Ref. 23, pp. VII-C-34ff)0
Continuous methods are described briefly by Wolf (Ref. 17).
Photovoltaic Storage Systems
Coal fired and nuclear steam electric plants can generate on arbitrary
duty cycles, and can track demand. Obviously solar generating stations can
produce power only during times of adequate insolation. Pumped hydro
storage and oil fired turbines for peak shaving are matters of utility econom-
ics and network distribution strategy. By contrast, in terrestrial solar
energy systems, some form of storage of backup power is an absolute necess
ity. A common utility system standard of reliability is one day of outage in
10 years. For an independent photovoltaic system to maintain that standard
of reliability, prodigious energy storage capacity would be required. The
alternative is to require backup power generating capacity from the utility
grid in addition to some manageable amount of storage capacity. Unfortu-
nately, therefore, while photovoltaic power may save fossil and nuclear fuel,
it would not reduce in the same proportion a utility's capital investment
(Refs. 9, 23 and 24).
Another option is to use the solar generating capabilities only for peak-
ing power (at less than full capacity). This would be realistic only in areas
where peak demand occurred during periods of maximum insolation (e.g.,
due to one-shift industry and air conditioning load). Environmentally and
economically this is not desirable. Economically this is not practicable and
environmentally it does not take complete advantage of all the "clean" solar
energy available.
Figure 3 (Ref. 23) displays the relationship between storage require-
ment and plant capacity factor for one application, imagined to require 12-
1/4 hours to 18 hours of storage during non-insolation periods. The amount
of fossil fuel backup is not specified. Figure 4 (Ref. 23) displays one candi-
date interfacing between photovoltaic generators and multiple storage modes.
Storage options are limited; some are summarized in Table 5 and
discussed in more detail below. A general review (somewhat dated) appears
in Ref. 25.
19
-------
<0
bO
INTERMEDIATE
0.2
0.4 0.6
Plant Capacity Factor
0.8
1.0
Figure 3. Storage requirements and associated plant
capacity factor for capacity displacement.
-------
UTIL1ZATIWI
Figure 4. Interfacing photovoltaic —fuel cycles (Ref. 23).
-------
TABLE 5. SUMMARY OF CANDIDATE STORAGE MODES
NO
NJ
Item
Minimum Economic
Size, Utility Appli-
cation
*
Estimated Costs
$/kW
Expected Service
Life (years)
Estimated
Efficiencies (%)
Use Fossil Fuels ?
Dispersed Capability?
Resource Limitations
r> c **
Keterences
Pumped
Hydro
10 MWh
200-300
50
65-70
No
No
Site
Limited
8, 26
Compressed
Air
200 MWh
80-700
20
45 Primary
70-75 Storage
Yes
No
Site
Limited
26
Batteries
10 MWh
180
10-20
70-80
No
Yes
Unknown
8, 19
Hydrogen
Cycle
10 MWh
75-350
30
50-60
No
Yes
None
27, 28, 29
Flywheels
10 MWh
100-400
30
80-95
No
Yes
None
8, 30, 31
Magnetic
104 MWh
500-600
30
90-95
No
No
Yes
1970 dollars.
<
For all modes, see Refs. 23, 25, 32
-------
1. Conventional pumped hydro storage. This is possible only
in terrain of substantial relief and ample surface water.
About 2500 to 5000 kWh/acre can be stored. There is no
prospect of significantly improving efficiency. Besides
large land commitment, there are very serious ecological
and aesthetic problems associated with draw down (Refs.
8, 23 and 26).
2. Underground compressed gas experience relates mostly
to helium and natural gas. Because auxiliary fossil
energy is used, about 4 kW h are recovered for every
3 kWgh in, but the real efficiency is 70 to 75% assuming
no gas losses through bedding planes and joints. Obviously
this can be used only where geological conditions are
favorable, or man made cavities can be exploited
(Refs. 23 and 26).
3. Batteries are becoming attractive at least for utility
application. The familiar Pb-acid battery is wasteful
of lead, which would come into tight supply. High
charging rates cause boiling and H^ evolution, and high
discharge rates cause electrode pitting.
However, newer high temperature storage cells using
Na, Li, CIL and S capable of high rate discharge are being
developed. The requirement that they be heated makes
them unsuitable for residential use, but for utility size
load leveling they may become very attractive. R&D
needs to be aimed at maximizing reliability at lowest
cost, even at reduced charge/discharge rate limits
(Refs. 8, 19 and 23).
4. Hydrogen cycles: water electrolysis, hydrogen storage
cryogenically or as hydride, and oxidation in H2~a-ir or
H2-Q, fuel cells. This is especially appropriate for
photovoltaic application because the dc power can be
used directly. Modular factory construction of com-
ponents allows add-on just as with the panels. Environ-
mental effects will be discussed later (Refs. 23, 27, 28
and 29).
5. Flywheels today are constructed of metal and/or glass
fiber and/or organic fibers. Coupled to a photovoltaic
generating system, a flywheel would be driven by a
variable speed motor-generator receiving its ac supply
from the system inverter. One design features a 100 to
200 ton rotor of 12 to 15 ft diameter rotating at 3500 rpm
with a storage capacity of 104 to 2 x lO'* kWh. It is cal-
culated that a 104 kWhr - 3000 kW unit would cost about
$325,000 and have an in-out efficiency of 93 to 95% if
maintained in H7 or He. These were 1973 dollars (Ref. 30).
Cj
The AiResearch Manufacturing Company of California
(Division of Garrett Corportion) has a U.S. Army
23
-------
contract to supply a 30 kWh flywheel. Their rotor has
a Kevlar™ rim, an aluminum hub, and an unspecified
web holding the two together. Total mass is 750 Ib,
yielding 40 Wh/lb. They state they could raise this to
— 55 Wh/lb; about twice what can be accomplished with
isotropics such as cast steel. By scaling to a cylindri-
cal configuration they feel they can achieve a 10 MWh
unit, about the minimum suitable for utility use. Their
unit operates in a vacuum of 1 fj. -I torr at a rim speed
corresponding to Mach 3, Discharge from storage is
accomplished by "dra-v down" to 1/2 velocity, correspond-
ing to 75% energy extraction, Maximum practical "draw
down" for a flywheel would be 94% extraction (4:1 velocity
reduction). Peripheral equipment costs limit extraction
(Ref. 31).
Another company in flywheel development work is
Rockwell International. They are working with composites
(Refs. 8, 23, 25, 30 and 31).
Power Conditioning and Handling
Photovoltaic receivers produce dc power which can be raised to 117
Vdc by appropriate series -parallel connections. Because American trans-
mission facilities and much of the load is designed for ac, inverters are
required. These now achieve 95 to 97% efficiency (Ref. 33). The basic
technology seems well developed, but there is some interest in further opti-
mizing subsystems (Ref. 19). Especially in smaller dedicated systems, the
economics are heavily dependent not only on storage but also on regulator/
inverter costs. At least some of the switches, breakers, transformers,
regulators, inverters, etc., may be dielectric liquid filled. This could be
transformer oils, PCBs or SF/.
Transmission lines will be required. In the West especially, large
amounts of electrical energy are wasted in long distance transmission.
There is very definite interest in conserving by grid dispersing solar gen-
erating plants nearer to load centers (Ret 16). This writer has not en-
countered any discussions of the effect this might have on the current trend
to very high voltage transmission. This should be investigated.
Central Station Photovoltaic Generation on Geosynchronous Satellites
This concept, reviewed recently by Williams (Ref. 34), has popular
appeal. Figure 5 depicts one conceptual design.
Although several aerospace companies and NASA laboratories have
proposed such stations in apparent seriousness and initial design studies have
been funded, capital costs are so enormous that our national priorities may
prevent construction of such an "installation," regardless of its merits. Table
6 presents one projection of satellite power station deployment.
24
-------
RECEIVING ANTENNA (6 X 6 MILES)
SOLAR COlt ECTOR (5X5 MILES)
- y
TRANSM'SS'ON
LINE
(2 MILES)
SUN
ROTARY JOIN IS
MICROWAVE ANTENNA
'(I X I MILE)
-CONTROL STATION
•WASTE MEAT RADIATOR
COOLING EQUIPMENT
Figure 5. A conceptual 10 MW satellite photovoltaic power station (Ref.9).
25
-------
TABLE 6. PROJECTED SOLAR POWER STATION DEPLOYMENT (Ref. 34)
NJ
Time
Period
1990-94
1995-99
2000-04
2005-09
2010-14
2015-19
Stations
Added
4
5
7
10
14
17
Shuttle Flights
per Year
1444
1805
2527
3610
4693
6137
Cumulative Costs,
Excluding R &D
($109)
68
149
261
419
635
893
Power
Delivered
(103 MW )
56
123
215
345
523
735
*
A fully reusable Shuttle with a 277,000 kg pa/load nuclear tug having 77,000 kg
payload from low Earth orbit to synchronous orbit.
-------
Features of this proposal which require attention are:
1. The use of large numbers of shuttle launches
(combustion of propellants)
2. The launching from earth of nuclear cargoes, sub-
sequently to be used in space propulsion (conse-
quences of an abort and burn-up)
3. The use of high intensity microwave beams to trans-
mit energy back to earth (physiological and ecological
effects)
4. The much reduced quantity of structural material, as
compared to terrestrial installations (mitigation of
resource depletion).
The geosynchronous satellite generating station has grave direct and
indirect environmental problem areas that will be discussed in the photo-
voltaic assessment section.
It is possible to propose a concentrator-thermal electric satellite power
plant; its problems are so similar to the photovoltaic concept that it will not
be given separate treatment.
Combined Receivers at Load Centers
Installations have been designed which combine a photovoltaic receiver
with a flat plate thermal collector, achieving a combined efficiency which is
quite high at no increase in area covered. See Wolf (Ref. 17) for a discussion.
Most buildings require both heating-cooling and electricity. By having
the solar photovoltaic array form the absorber surface of the thermal col-
lector, component cost is reduced by multiple use, and up to 60% of the total
available solar energy is absorbed (Figures 6 and 7).
This system does not appear to have any adverse direct or indirect
environmental effects not discussed in the photovoltaic and flat plate sections.
On the beneficial side, there is a real saving in land or building area, and
some material saving. No further discussion of this concept seems to be
required now.
Concentrator Photovoltaic Systems
GaAs is more temperature resistant than other photovoltaic semicon-
ductors. Consequently, it is possible to place this photovoltaic material at
the focus of concentrator mirrors of various configurations. Substantial
reductions in GaAs requirements result — roughly by the same factor as the
concentration ratio. One thousandfold savings are conceivable with this
scheme (Ref. 11). The environmental consequences of this kind of installa-
tion will be discussed in the concentrator chapter and in the photovoltaic
materials resource section.
27
-------
Burn ionr
.;.,^.-ANII in i ircTioN
— StCONfGLASS
^jl^
"I"!1 ^L/LM*— SOLAR CELLS
= "^?79
HEAT
TRANSrEB
LIQUID OR "
GAS
DUCTFOFI
HCATTRANSFER
LIQUID OH GAS
THEFIMAL
INSULATION
Photovoltaic solar array can be mounted inside a thermal flat plate collector
for combination heating and photovoltaic power systems.
Figure 6. Photovoltaic-thermal combined collector (Ref. 17)
(Reproduced by permission of author).
/• — ~v Fossi -Fired
Q£_ f Fossi l\ Bock Up
^_ I Fuel J *" Electric
T ^ *S Generator
Bui Iding
Load
/ \ \ r -t n
\\X*^ D'<"' I ~ 1 i Electrico / \
\Y'/\\ ei-tricl'X ( ^ Battery f fc D.C. to A.C. ' Energy
_ X/ l V bloroge Inverter ' — ~
Combined
Photovoltaic ' Spocc Con
and Thermal Ener9>'
Solor Collector Thermal
1 Cllily),
V ^_ Waste Heat Heotina and P:r;n^
•Unrnnr Cooling Equipment *" Network
1 » ' ' 1 '
1 '
L J
s* N. Fossil - Fired
/^Fossil A Bock Up
V Fuel I Space Conditioning
^ — ^ Equipment
D D
Jitioning
*^ LJ | | LJ
Figure 7. Combination photovoltaic and heating-cooling system (Ref. 17)
(Reproduced by permission of author).
23
-------
ENVIRONMENTAL ASSESSMENT -PHOTOVOLTAIC SYSTEMS
Siting and Grid Dispersal
Most writers on the subject assume that major utility solar installa-
tions will be located in the U. S. Southwest, in areas of highest insolation.
It is important to distinguish between total insolation and direct insolation.
Concentrator systems, including photovoltaic, can use only the latter
(Figure 8).
200
Figure 8.
Isopleths of mean daily direct solar radiation, Langleys/
Day (Ret 35) (1 Langley = 1 cal/cm2 = 3.62 Btu/ft2)
Flat plate systems, including photovoltaic panels, use both direct and diffuse
radiation (Figure 9). These figures reflect both the solar radiant intensity
reaching earth at each latitude, as well as meteorological effects such as
cloud cover. They may not contain corrections for interruptions due to dust
storms.
Season-to-season variations cannot be leveled by storage. In regions
where season-to-season variations are large (Table 7), non-solar generating
facilities must be available during the low-insolation months.
Grosskreutz (Ref. 35) has described criteria and procedures for siting
solar steam-electric plants. Aside from concern for water availability, his
study should apply to photovoltaic panel systems. In addition to several
legal, social, and institutional matters, his criteria include:
29
-------
TABLE 7. SOLAR RADIATION AT SELECTED LOCATIONS
IN THE UNITED STATES DURING 1970*
Average
Location
Sea ttle-Ta coma,
Washington
Fresno,
California
Tucson,
Arizona
Omaha,
Nebraska
San Antonio,
Texas
Lakeland,
Florida
Atlanta,
Georgia
Burlington,
Vermont
Jan
278
710
1110
111
862
1029
873
581
Feb
688
1117
1391
1110
1103
1436
1203
781
Mar
1069
1709
1750
1284
1432
1480
1288
1088
Total
Apr
1354
2205
2202
1576
1506
1983
1635
1384
Daily
May
1950
2609
2435
1939
1906
2079
1991
1447
Insolation (Btu's per square foot per day)''""
Juri
2065
2579
2449
2165
2083
2042
1854
1758
Jul
2105
2576
2190
2002
2176
1883
1917
1587
Aug
1750
2412
1983
1865
2057
1680
1628
1835
Sep
1217
2050
1735
1280
1587
1639
1591
1195
Oct
747
1425
1587
944
1388
1436
1021
759
Nov
370
910
1221
581
1310
1302
955
444
Dec
229
614
870
596
784
1169
714
448
Annual
Average
1152
1743
1745
1351
1516
1597
1389
1109
Source: Commerce, 1970: Vol. 21, Nos. 1-12.
'" ? ?
"lOOO Btu/ft -day = 276.24 Langley/day. 1 Langley = 1 cal/cm .
-------
Figure 9. Distribution of solar energy onto a horizontal surface.
(figures give solar heat in Btu/ft per average day;
1000 Btu/ft3 = 276.24 Langleys) (Ref. 9).
• "Avoid dry lake playas, depressions, sand dunes, and areas
of uncompacted sand"
• "Avoid seismic faults (this would be much less important in
these applications, not requiring "power towers.")"
• "Avoid proximity to airports and flight corridors." (His con-
cern is not explained. Presumably highly reflective surfaces
covering many square miles could be visual flight hazards
to flight personnel, but perhaps no more so than large
bodies of water.)
• "Seek relatively flat areas with good drainage. Slopes which
face south are acceptable, especially for central receiver
system configurations." (It is not clear why flat panel arrays
cannot be placed on slopes with equal ease. To reduce sub-
sequent erosion of terrain disturbed during construction,
however, slopes should be avoided.)
• "Access. .... with a minimum of secondary road improve-
ment or spur construction" (Almost unique among power
plants, most, of the construction of these installations will
be modular, with fabrication occurring in remote factories.
Construction will place unusually heavy demands upon
surface transportation.)
31
-------
Several writers have emphasized that the insolation, meteorological and
terrain criteria will necessarily force siting of solar power stations in
regions of desert biome and uniform ecological characteristics.
An important feature of photovoltaic siting relates to the very small
economies of scale (Ref. 5). This means that the size of a unit may be
decided by economics of peripheral equipment, transmission equipment,
access preparation, etc.
An EPRI spokesman (Ref. 14) has addressed the security aspects of
grid dispersed stations —physical security is a problem in grid dispersed
systems, but the problem can be solved. System security favors dispersed
installations. He too emphasized that reduced transmission and distribution
costs, reduced capital investment, and reduced peak demand problems all
favor serious consideration of grid dispersed solar power stations,,
The Auburn University Summer Fellows program final report "MEG-
ASTAR," presents some interesting observations on siting related to insti-
tutional and social matters, which can be quoted here as usefully as later
(Ref. 36).
"The diffuse nature of solar energy allows the generation of solar
power to be widely distributed. The 100 GW capacity suggested
for the year 2000 in the previous section would involve roughly
450 square miles, but this will not be sited on a single 21 mile
square. There are several advantages to a distributed source.
Transmission line losses are reduced, the need for more trans-
mission line corridors is eliminated, and individual plant
failures are felt only locally. Also, there may be a positive
social impact associated with decentralized power sources.
The tax and cost structure and responsibility associated with
smaller, local power plants provide a healthy social climate
and a feeling of "oneness," belonging and pride in the com-
munity. The size distribution of American cities shows a
great many communities of 10,000 to 13,000. If average solar
insolation figures are used, a solar plant (photovoltaic or
thermal) of about 0.1 square miles and some form of storage
(batteries, flywheels, electrolysis/fuel cells or thermal)
could supply all the electrical power needs for a community
of 12,000. This corresponds to an average power demand of
about 7 MW day and night, summer and winter. Clean air
is especially desirable in residential areas; pollution from
power generation would be completely eliminated. In defense
of the existing centralized utility operation it would be noted
that operating costs usually go up with decentralization. On
the other hand savings in transmission line and conventional
fuel costs may off set the higher operating costs."
Szego (Ref. 26) states that transmission costs account for about one
third of our electricity costs. This suggests that very careful analysis needs
to be applied to the economics of grid dispersed stations near load centers
in the Southwest. Such dispersal of solar generating capacity would also
32
-------
reduce transmission losses and thus reduce dependence on fossil and
nuclear fuels in some degree.
Szego has also commented on the urgency of developing siting proce-
dures, and related matters:
"Siting a Major Stumbling Block. Siting decision methodology
must promptly be developed. The struggle invariably joined
relative to siting of power plants or pumped storage facilities
has become reflex rather than a rationally motivated, selective
opposition-protagonist dialog. Ecologically-motivated re-
sistance to siting selections is inevitably seriously counter-
productive. The longer a plant is delayed, the more the
electrical utility industry turns to gas turbines, which have
lower efficiency, and thus create about 75 percent more
thermal and gaseous pollution than larger steam plants, and
use more fossil fuel in the same proportion. This is not
the goal of the siting resistors, but it is achieved nevertheless."
Direct Effects
Surface Water Effects and Thermal Pollution: In much of the South-
west, especially, surface water and ground water are in exceedingly short
supply. This makes particularly cogent the requirement that the quality of
U. S. waters not be degraded and that States and Federal facilities have
water quality management plans (Ref. 37).
An outstanding virtue of photovoltaic panel receiver systems is their
trivial water requirement:
• No thermal pollution occurs because no cooling water
is needed.
• Because no cooling towers are needed there are no
problems with biocides, slimicides, detergents, anti-
corrosion agents, etc., in blowdown or once-through
waters.
• At least at present, photovoltaic panels are hermetically
sealed. This avoids discharge of eroded CdS or other
toxic semiconductor materials onto the land and into
the surface waters under normal operation.
• It is still uncertain whether dust accumulations on panel
surfaces will create unacceptable transmittance losses.
Pilot studies have not often encountered serious trans-
mittance problems ascribed to dust. Possibly periodic
or occasional washing will be needed. This would have
to be done in situ using some kind of automated down-
wash system, and on a batch work basis to avoid excessive
water demand. We have no information relative to
possible detergent demand in this operation.
33
-------
Large areas of the Southwest have cavernous limestone
bedrock (Figure 10)., A feature of such terrain is its ability to transport
Ground-water:
pollutants unexpectedly and rapidly to remote areas by direct conduit trans-
port in the vadose groundwater (Ref0 38).
Figure 10. Principle known regions of cavernous limestones
in the contiguous U.S.
Initial land grading could conceivably disturb the subsurface geohydro-
logy. Although the desert areas do not receive much rainfall per unit area,
very large areas will be involved.
If the recharge area for an existing vadose groundwater system were
almost wholly under such a disturbed area, distant aquifers could be affected.
Figure 11 displays the relative paucity of usable aquifers in the Southwest.
Extreme care must be used to avoid degrading in any way the quality and the
availability of the waters of the Southwest. This will be far more difficult in
the case of steam-electric concentrator systems, discussed later.
New towns will arise in rather inhospitable areas to accommodate con-
struction and operating personnel. These communities will require potable
water, sewerage, and solid waste management services. Unusually careful
regional planning will be required in order to supply these needs without
infringing on appropriated surface waters and degrading surface and subsurface
34
-------
Patterns show areas underlain by aquifers generally capable, of
yielding to individual wells 50 g. p. m. or mart of water containing
not matt than 2,000 f>. p. m. of dissolved solids (includes some areas
where more highly mineralized wattr is actually used)
Watercourses in which ground water can in replenished
• by perennial streams
Buried oaf leys not now occupied by perennial streams
gj Vnconsolidaifd and semuonsolidattd aquifers
aquifers
gjjjj§jj§j Both unconsolidaUd and consolidated-rock aquifers
f j Met Inown to be underlain by aquifers that wilt
gtnerally yitld as much as SO g. p. m. to wells
Figure 11. Ground-water areas in the U.S. (Ref. 39).
35
-------
waters. ERDA is not yet addressing water problems with much intensity
(Ref. 12).
Grosskreutz (Ref. 35) has suggested that water might be brought to
these developments in canals or pipelines. He is almost alone in alluding
to the effects of groundwater pumpage on aquifer drawdown. Even if sup-
porting communities are supplied by well water, it does not seem likely that
their demands will be large enough to cause surface subsidence. That
possibility should be considered by site selection teams, however.
Land Use: Among direct effects of a major solar electric installation,
land use related effects are certainly the most obvious, if not necessarily
the most adverse.
The upper theoretical limit for solar electric conversion and land use
efficiency would correspond to about 2 mi2/GWe (Ref- 36)- To expect a utility
to achieve this power generation "density" would be unrealistic. In any event,
land acquisition costs may be too trivial a fraction of installed cost to pro-
vide much incentive for stringent economics in land use. Cost of construction
and maintenance argue for reduced area density of receiving panels. More
realistically we should expect a 1 GWe photovoltaic plant to occupy 4.5 to 17
mi2 (Refs. 7, 23 and 36). The AEC figure (Ref. 9) of 30 miYGWe seems
spurious and a figure of 660 mi2/GWe (Refs. 23 and 36) which has been per-
petuated seems to be due to an improper extrapolation of data for residential
systems.
Referring to Table A-l, we find a year 2000 Scenario III projection of
50 GWe solar-electric. Assume 40% of this is photovoltaic; if we accept the
Project Independence area density value of 7,,7 mi2/! GWp^ (Ref. 23), we
project a land use of 154 mi . This would not occupy a single 12.4 x 12.4 mi
site (cf. our earlier discussion on grid dispersal). Project Independence
(Ref. 23) projections are more optimistic than ERDA-48 projections, sug-
gesting photovoltaic power systems could supply about 191 GWpk in a
"business as usual" schedule and 1100 GWp^ in an accelerated schedule.
Table 8 summarizes some land use data.
TABLE 8. YEAR 2000 PHOTOVOLTAIC POWER CAPACITY PROJECTIONS
AND CORRESPONDING LAND COMMITMENT
Forecast
ERDA-48* Scenario
III x 40%
Project Independence**
Business as Usual
Project Independence**
Accelerated
Total
Capacity
(GWe)
20
191
1100
Land Area
(mi2)
154
1473
8492
Total Contiguous
U.S. Land Area
(%)
0.005
0.05
0.27
*!Ref- !•
**Ref. 23.
36
-------
The densities used above are significantly less than the densities
theoretically possible. This is not only because production panels cannot
perform to laboratory standards, but also because in a real installation the
panels would not be planted side-by-side in tile fashion. First, they would
be inclined at the latitude angle above horizontal or perhaps a few degrees
more (Brownsville, Texas is about 19°30'N, Albuquerque, New Mexico,
about 35°N, and Boulder, Colorado, exactly 40°N). Furthermore, space is
needed for: (1) construction and service roads; (2) transmission lines; (3)
power conditioning facilities; (4) energy storage facilities; (5) substations;
(6) maintenance buildings; (7) control room and office buildings, and (8)
access roads and parking lots, etc.
Many of these auxiliary facilities will have to be distributed throughout
the field of panels. To avoid shadowing the receiving surfaces, the auxili-
aries will have to be low. For example, perhaps transmission lines would
be in trenches, and power conditioning and storage modules in low block
buildings in small islands distributed throughout the field. Manway corri-
dors will be needed between rows of panels; at least every second manway
will have to be at least wide enough for a pickup truck. Near term cost
tradeoffs will dictate even greater spacing to avoid self shading. Thus,
perhaps the visual appearance of such an installation would approximate a
large and orderly storage yard.
To help put these area considerations into perspective: all the electri-
cal power used in the U.S. in 1972 could have been generated in 3000 mi in
Arizona at a generating efficiency of 12% (Ref0 24). The 2.1 GWe "Four
Corners" coal fired powerplant of Arizona Public Service is reported to have
already under lease for coal stripping more land than would be needed to
construct a 1 GWe solar plant using 6.2 mi^ of collector surface (Ref. 7).
Figure 12 quantifies this. It is adapted from an AEC comparison
(Ref. 9), which had claimed a 1 GWe solar electric plant would require
30 mi of land. This was in the solar assessment in the "alternatives con-
sidered" portion of the LMFBR proposed final E. LS. The AEC "Range for
Coal Surface Mined" has not been altered. The accompanying text does not
elaborate on the input assumptions, or whether indirect land disturbances
were included. The comparison looks extremely favorable to solar develop-
ment, even granting that some strip mines are being reclaimed.
Visual Effects: The appearance of a photovoltaic plant has been de-
scribed earlier. Environmentalists who enjoy the wildness and openness of
the desert would probably prefer no development. At the same time, excel-
lent arguments have been advanced that among the few obvious alternatives
is more coal mining (Ref. 40). Solar electric development appears to be less
destructive of aesthetic values than coal mining and preparation, without the
directly identifiable air and water pollution impacts. Further, photovoltaic
plants will have no high structures other than departing transmission lines,
common to all electric generating facilities.
Terrain Effects: This subject appears to have two aspects: terrain
modification during construction, and during operation. Additional indirect
effects will be discussed in a later section of this report.
37
-------
8,000
p ! 6,000
O
o
LJ
CD
CC
ID
h-
-------
Construction Phase: Most conceptual designs envision
photovoltaic receiving panels to be placed on level or
gently sloping terrain. Grosskreutz (Ref. 35) recom-
mended this in his siting study. Candidate sites are
unlikely to be forested. Consequently, only minimal
grading and land clearing are likely.
Many of the candidate sites are in desert areas in
which arroyos and dry •washes are common. Most
especially in sites near mountains, these should not
be obstructed. Plant layout must be preceded by an
understanding of surface hydrology after storms in
the mountain watersheds supplying the arroyos.
Solar panel supports presumably will require footings
to be poured, but no foundation excavations. There
will be a few buildings of conventional construction,
and heavy peripheral equipment which will need to be
supported by concrete pads.
Operating Phase: There is a potential for water run-
off to cause erosion. Panels or arrays of panels of
perhaps as much as 100 m^ individually, will be
inclined at perhaps 35deg above horizontal. During
infrequent rainfalls, water will run off these im-
pervious surfaces and could strike the ground with
substantial velocity. Some provision must be made
to decelerate and distribute run-off water onto the
ground to avoid erosion.
Agriculture and Terrestrial Ecology: The following passages are
quoted from Ref0 9:
"The land now most likely to be used for central-station thermal
conversion or photovoltaic solar plants is in the desert and semi-
desert Southwest. Some open-range grazing is done on parts
that receive a little rainfall. The vegetation is so sparse
(chiefly creosote bush and white bur sage, with lesser amounts
of saltbush paloverde, catclaw, and cactus) that the land is
not very valuable as graze land (Ref. 41 )„ During the winter
and summer rains, the desert usually has a lush cover of
annual grasses and forbs and is valuable for grazing for a
short period. This land is used for very little human habita-
tion or industrial activity except in the cities. If solar power
plants were developed, the economic value of adjacent land
would undoubtedly increase but aesthetic values might be
reduced.
"To develop land for large solar power plants will necessitate
the construction of roads and sites for the solar collectors.
39
-------
This process may involve destruction of much of the local
ecosystems. On the other hand, some persons believe that
if half the land were shaded, it could be greatly improved
as graze land. This modification would require new plantings
and management. Before a judgment could be made on the
feasibility of this agricultural use of the land, agricultural
research would be necessary.
"Shade itself would have a significant effect on the vegetation.
Plants indigenous to the desert require high-intensity sunlight
Some ecologists believe that these plants would die out in
shaded areas (Ref, 41).
"Many species of mammals, reptiles, amphibians, birds and
invertebrates are in these desert regions. Some are
threatened species. The construction of large central station
solar plants might upset the ecological equilibrium, but to
predict the changes that would occur in animal populations
is difficult. The extent of change would obviously depend
on the number and size of plants constructed."
Clearly there is some potential for multiple use, which can be evalu-
ated by research. The development we are discussing is to take place over
several decades, and very slowly at first. Ample opportunity exists to
perform the agricultural and ecological studies needed.
Evaluations of ecological effects must be weighed against known effects
of fossil fuel fired power plant emissions. The latter knowledge is also at
a rather primitive stage. Most of the studies of vegetation injury due to air
pollution, for example, were performed in the North, the East and the
Southeast. These studies are not relevant to the ecologies of Southwestern
deserts or high arid plateaus. Our knowledge of the environmental impact
of conventional generating facilities in these ecosystems needs to be upgraded.
Even back-up facilities (perhaps older less efficient coal fired plants which
would otherwise have been retired), will produce SO and atmospheric par-
ticulates in an ecologically signficiant quantity.
Air Pollution: No potential for air pollutant emissions from photo-
voltaic panels seems to exist. Environmental effects of storage options are
discussed in a later subsection.
Power conditioning and other peripheral equipment may contain PCBs
or SF/. SF/ is non-toxic, but had some potential for use as a sensitive
spiking agent for plume tracing (Ref. 42). This application may be eliminated
if accidential releases continue to occur.
PCBs are known toxic agents which concentrate in aquatic organisms
and are transmitted up food chains. Their fates and ecological effects in
desert biome should be determined. Possible atmospheric releases of PCBs
must be prevented during fabrication, installation, normal operation, and
maintenance activity.
40
-------
Transmission Line Effects: There have been reports that electro-
magnetic radiation from power lines can affect animal growth (Ref. 43) and
can affect the earth's magnetosphere (Ref. 44). Atmospheric ozone is
commonly thought to be noted near high voltage transmission lines.
These effects, if real, surely must be independent of mode of power
generation. Further, to the extent that solar generators are grid dispersed,
lower transmission line voltages will be practical. It seems probable that
solar energy development has some potential for reducing such effects below
what could occur if conventional coal or nuclear fueled plants were intro-
duced in the same general regions. Conversely, intensive solar develop-
ment would introduce such effects into some specific desert locales not
previously exposed. Field research would be needed to establish the reality
and extent of the effects, if any, and the mitigating technology, if required.
Visual effects, and right-of-way construction effects, may be very
severe, as with any utility network installation. The severity of these trans-
mission line impacts, however; cannot be evaluated effectively in a non-
site-specific study.
Weather Modification —Direct and Indirect—All Technologies: Coal
and nuclear fueled power plants are only 30 to 40% efficient. As a conse-
quence, prodigious amounts of reject heat are released to earth's atmos-
phere by steam electric generating plants and heating plants. The Project
Independence Task Force examined this problem in great detail (Ref. 23,
pp. A-II-iff) primarily on a global scale. The task force conclusions are
that human influences on the global heat balance have so far arisen from
five major sources, of which three are related to fossil fuel combustion:
1. Particulate Emissions
Fuel combustion, industrial activities and natural sources
combine to emit atmospheric particulates which probably
produce a net cooling by reflecting or scattering sunlight
away from earth.
2. Carbon Dioxide Emissions
Fossil fuel combustion has created a 9% CO-, increase since
1870, and will produce a projected 30% CO2 increase by year
2000. Atmospheric COo produces a net heating effect by
absorption of solar radiation.
3. Direct Release of Heat
They state that "COz emissions appear to be the most likely effect to cause
global climate changes in the long run."
#
Much of this discussion would be appropriate in the section on indirect
effects. However, it seemed convenient to consolidate the discussion.
Likewise, concentrator and flat plate effects are included here.
41
-------
To the extent that solar power generation displaces fossil fuel com-
bustion, the developing industry will have a beneficial effect in decelerating
our inadvertent climate modification. The task force position, written by
R. S. Greeley of MITRE Corporation (Ref. 23 loc.cit.) is aptly summarized:
"The use of fossil, nuclear and, to a lesser extent
geothermal energy sources throughout the world at
rates possible to be attained before the year 2100
could cause an increase in average global tempera-
ture. The size and effects of such a temperature
increase have not yet been calculated with any
degree of certainty.
"The use of solar energy instead of "stored" energy
sources would avoid the problems of waste heat re-
lease and carbon dioxide emissions and thus avoid
the possibility of such a global temperature increase.
However, the large areas needed for solar energy
collectors could pose a problem in land use and
could affect climate through changes in energy
distribution patterns.
"More extensive measurements and calculations of
global climate changes are needed in order to under-
stand the causes of such changes, particularly from
human activities, and to predict their impact on
the earth and on human society." (Emphasis added.)
The impact of solar collectors on heat balance is summarized sche-
matically in Figure 13. The position adopted by many authors is that al-
most the same quantity of solar energy is returned to earth's atmosphere
and to space, with, as without, solar collection, but that the geographical
location of such return is altered. This is an oversimplification. A signif-
icant fraction of man's energy use is dedicated to unreversed enthalpy in-
crease — for example, in reduction of bauxite to produce metallic alumi-
num. This is a very electric energy intensive industry. However, the point
is valid that fossil fuel combustion and nuclear fission release sto'red energy
not previously part of earth's heat balance in this geological era.
Because solar plants are likely to be distributed, we are unlikely to
create giant "heat bubbles" over thousands of square miles of consolidated
solar collector. Because solar development is starting out small, we have
an opportunity to learn how serious heat balance effects may (or may not)
be before we are irrevocably committed to enormous consolidated projects.
Solar thermal (steam electric) plants share with coal and nuclear
plants an efficiency of about 30 to 40%. The reject heat is likely to be dis-
charged in cooling towers. Sites in the desert southwest may not have
adequate unappropriated water; dry cooling towers may be necessary. If
wet towers are used, drift and fog will also result, but will dissipate more
rapidly than in, say, Baltimore. (Cooling tower chemicals will be released
to a fragile ecology with blowdown, fog, and drift.) As a consequence, solar
42
-------
NORMAL
SOLAR ENERGY COLLECTION
UJ
Figure 13. Thermal energy balance (Ref. 7).
-------
steam electric plants will create the same kind and quantity of atmospheric
thermal effects as the equivalent conventional plants.
The potential for local weather modification due to the artifically
altered humidity needs to be investigated. It is possible also to imagine
some rather unusual flora developing in the "moist desert" created adjacent
to a cooling tower] Furthermore, the utility companies have a vested
interest in learning about local weather modification, because the per-
formance of the solar collectors or concentrators is related to cloud cover,
light scattering by haze, etc.
Urban roof top and sidewall flat plate collectors may be expected to
alter the heat balance of a city if they become common features. The extent
to which they will alter the urban "heat islands" or "bubbles," and there-
fore the cloud cover, precipitation, mixing depth, and frequency of stable
inversions is apparently unknown. This effect needs to be studied; it seems
capable of computer simulation, at least with quite simple urban models.
Probably some useful insights could be gained in laboratory modeling, also.
The entire complex of interfaced effects related to altered vertical
temperature gradients is likely to depend on several geometrical and mete-
orological parameters in a feed-back loop sufficiently subtle as to preclude
prediction prior to well designed research.
Among these parameters we have to include roughness. Advection,
mean winds, and turbulence in urban environments are extraordinarily com-
plex problems. Researchers in this area have found that laboratory simu-
lation in meteorological wind tunnels is often necessary (Ref. 45), and
frequently successful.
In Figure 14 the boundary layer effect of modest roughness is dis-
played. A large concentrator installation with many power tower s of 200 to 300
meters height, and many heliostats projecting several meters above the
mean ground plane might approximate a woods or small town in its meteor-
ological effect. Cermak (Ref. 46) believes there is some potential for
problems if large areas are used. He believes research here could be a
signficiant aid. He wonders about effects on downwind urban areas. The
problem should be amenable to both model studies and computer simulation.
Cermak suggests examining the internal boundary layer and on up, studying
the effects of stepwise changes. He feels there is some potential for opti-
mizing solar receiver field and structure configurations. A substantial
applicable literature now exists in both the hydraulic and the meteorological
literature.
Auer, at Laramie, has been participating in the METROMEX studies
in the St. Louis area. Among his observations (Ref. 47):
• Most meteorological anomalies are associated with land use
changes.
• When vegetation cover is <_5%, or when there is an 80 to 90%
surface change, the meteorological changes occur not just at
the surface, but up through the mixing layer.
44
-------
Velocity Profiles over Terrain with
Different Roughness Characteristics
for Uniform Gradient Wind Velocity
of 45 m/sec
500 -
400-
300-
200-
100-
GRADIENT WIND 45 m/sec
ROUGH WOODED COUNTRY,
TOWNS, CITY OUTSKIRTS
FLAT OPEN COUNTRY,
OPEN FLAT COASTAL BELTS
Figure 14. Effect of roughness on boundary layer velocity (Ref. 23).
45
-------
• Changed evapotranspiration leads to changes in cloud
cover.
« "Potential temperature" is higher over natural than
urban areas.
These observations suggest that large scale surface alterations associated
with installation of solar photovoltaic receiver or heliostat fields need to be
examined by meteorologists. He suggested that one might expect some
effect in altered thunderstorm precipitation systems.
Auer suggested however that land use occasioned meteorological
anomalies in the Southwest would probably not be as severe as in the Mid-
west or Northern plateau areas, for example. This is because, in the
Southwest, there are fewer precipitation systems. The greatest land-use-
related meteorological changes are associated with areas displaying the
largest normal changes.
Auer concluded by noting a caution: that much of this is surmise.
Nevertheless, like Cermak, he felt that this is possibly an area that merits
examination.
We have been discussing alterations in local weather. One must not
assume reflexively that these effects are necessarily adverse. From the
multiple-use point of view, agriculturalists might find some small changes
to be beneficial. We have already noted the suggestion that shaded areas
beneath panels might be put to constructive use. Several writers, including
the authors of the LMFBR E.LS. (Ref. 9) suggested that increased surface
roughness would have a beneficial effect in reducing surface wind erosion.
Severe hailstorms, in "hail alley" particularly (Colorado-Nebraska
border vicinity), cause $600 million agricultural damage and $150 million
property damage annually (Ref. 48). This damage is severe —and worse,
it is concentrated (Ref. 49), so that the economic impact of a specific storm
strikes with extraordinary severity within one county, or even one section.
This has two implications for solar development.
First, the areas struck by most of these hailstorms are not now high
technology areas containing great capital value per unit area. What would
be the economic and generating capacity loss if such an onslaught chanced
to strike a photovoltaic generating station?
Second, to what extent could the frequency of such episodes be modified
inadvertently by changes in heat balance, humidity or boundary layer
velocities ?
Noise Pollution —All Technologies: Photovoltaic generating stations
will be quiet. Presumably there will be the usual 60 cycle hum associated
with transmission lines.
Flat plate solar collectors will probably require pumps or blowers to
circulate heat exchange fluids. This equipment, even though situated at the
load center, need be no more annoying than the sounds of forced draft
46
-------
heating and air conditioning equipment. Further, it will not necessarily
be located in the circulating building air.
Concentrator steam electric systems will require some of the same
rotating equipment used in conventional installations. Thus, one may expect
such a solar facility to be no quieter at the equipment than an oil fired tur-
bine generating plant for example. However, sound levels at the perimeter
fence may be very low. Further, solar installations are not expected to be
sited in populated areas.
Radiation —All Technologies: No nuclear materials are associated
with any terrestrial photovoltaic installation. No ionizing radiation of any
sort can be expected to be emitted from solar facilities.
The geosynchronous satellite generating station however is quite a
different matter. It is proposed to use nuclear tugs, for example. At earth
launch, the nuclear materials would be cargo. However, the consequences
of an aborted launch followed by burn-up and/or disintegration are staggering
to contemplate. Most careful consideration must be given all design options
and all abnormal event scenarios prior to any construction and operation
commitment.
It is proposed to transmit generated power back to terrestrial receivers
using microwave radiation at about 3 x 1(K Hz. Entirely aside from
"leakage" around the mainbeam, one must wonder about the consequences
of incorrect attitudes. Would it be possible for transient difficulties in
attitude correction to cause the main microwave beam to strike urban areas?
And what would be the consequences to aircraft passengers intercepting the
beam?
The physiological and ecological effects of microwave radiation are
certainly not thoroughly understood. Enough is known, however, to suggest
great caution (Refs. 7, 8 and 34)0
It would be very surprising if there were no public concern over these
potential problems (whether well founded or not). In fact, it seems quite
possible that public reaction might become at least as vigorous as that
surrounding nuclear plant siting.
Historical, Archaeological, and Paleontological Values: Very little
can be said on these matters on an a priori basis, because they are com-
pletely site-specific. Because solar development will start slowly, and
because of the light construction occasioned by most photovoltaic solar
facilities, salvage archaeology and paleontology can perhaps precede and
accompany construction. Many of the early Indian settlements of the South-
west (such as Canyon deChelly and Mesa Verde) were located in or near
rough terrain unsuitable for solar station installations.
Accidents, Disasters and Sabotage: This subject appears to have two
aspects: the inherent vulnerability of central station and grid dispersed
installations, and the environmental consequences of damage:
47
-------
• Vulnerability to Accident: We have already discussed hail-
storms. Photovoltaic panels may be especially vulnerable
to fracture by hail or airborne debris.
It is possible that polycrystalline CdS panels could continue
in operation after damage from impact. Single crystal Si
panels may not survive. Although a catastrophic event
effecting many panels is conceivable (e.g., hail), it is not
likely.
The authors of the Dow Process Heat Study (Ref,, 50) have
considered the reliability-vulnerability issue from the plant
operations point of view. They suggest that panels "quite
probably. „ . could be maintained on a replacement basis
with a malfunctioning unit removed and spare inserted."
Near term capital costs of photovoltaic panels are likely
to be so high as to preclude a large inventory. Heliostat
pieces may be easier to stock..
• Vulnerability to Sabotage: Szego (Ref. 26) reports that
InterTechnology Corporation has examined various
vulnerability and low cost countermeasure scenarios.
He states that qualified requestors may obtain further
details from ITC.
On an a priori basis we see that a large field of solar
receiving panels might be vulnerable to terrorist attack.
However, the diffuseness and modular construction
preclude wide-spread or irreparable damage, at least
in a photovoltaic facility.
In this respect, a photovoltaic solar installation offers
very favorable alternatives to nuclear, or even coal-
fired plants. Further, a solar installation is more
likely to enjoy widespread public acceptance.
As Zeren (Ref. 14) indicated, the utility industry is
aware of the increased vulnerability of grid dispersed
stations, but feels the compensating advantages may
outweigh the objections to grid dispersal (cf. "Siting
and Grid Dispersal "). Simple acts of vandalism
are more possible in such small sta.tions, as well
as in residential installations. Prince (Ref. 6)
states this has not been a problem to date.
• Seismic Events: Although optimized orientation of
photovoltaic panels is desirable, precise alignment
is not absolutely essential for extraction of power at
derated capacity. Light construction favors prompt
repair of damaged supports. Modular construction
permits portions of an installation to be out of service
without scramming the entire plant. The greater part
48
-------
of a photovoltaic plant seems to be less easily
damaged than a conventional plant.
Power conditioning, storage, and transmission
peripherals would be about as vulnerable as
equivalent facilities at conventional plants,
• Fire: There seems to be little prospect for a fire
to begin, or to propagate. An important exception
is in the case of hydrogen cycle energy storage.
Modular construction reduces the impact of such
an accident (Ref. 27).
• Environmental Consequences of Damage: In a photo-
voltaic installation, the airborne distribution of CdS
or GaAs and its introduction into surface waters
appear to be the only non-local environmental effects
of the damage scenarios we have addressed above.
With respect to surface water contamination—it
appears to be a trivial engineering problem to design
panels to preclude this in the case of simple fracture.
Normally CdS panels would be hermetically sealed.
If the transparent seal were fractured, erosion would
be possible. By judicious design, perhaps at some
loss in efficiency, erosion products could be captured
within the panel behind the remaining panel seal.
In the case of extensive damage such as could be
caused by a tornado, for example, rapidly mounted
emergency procedures to collect and secure damaged
panels would be required.
Geosynchronous Satellite Generating Station — Environmental Problems
Propellant Combustion: One concept of a satellite power station was de-
picted in Figure 5. Other conceptual studies are to be performed under two
18-month NASA contracts recently announced (Ref. 56) to be in final negota-
tion with Grumman Aerospace Corp., Bethpage, N. Y., and McDonnell
Douglas Astronautics, Huntington Beach, Calif. NASA's Marshall Space
Flight Center, Huntsville, Ala., and Johnson Space Center, Houston, Texas,
are committed to space station projects which are to incorporate solar
generating stations.
These stations are to be located at 22,300 mi altitude and house crews
of about 200. Typical concepts involve construction of solar power satellites
by these space station crews. Satellite solar panel "fields" are currently
believed to require structures 20 to 50 mi^ in area, and perhaps 600 ft in
depth. One concept envisions use of discarded fuel tanks as structural
elements. Target date for design decision is now in 1978.
49
-------
Current plans require lifting of crews and very large quantities of
materials into orbit using the space shuttle launch vehicle (SSLV). Williams
(Ref. 34 and Table 6) projects 1805 SSLV launches per year in the period
1995-1999.
Each SSLV launch requires combustion of 1,234,000 Ib of liquid fuel
(H2 + O2) producing approximately that quantity of water vapor, and potenti-
ally vast quantities of NOX due to reactions in entrained air.
In addition, each SSLV launch requires combustion of about 2,2 x 10"
Ib of solid fuel (Ref. 57). A typical solid fuel composition is given in Table
9 (Ref. 58). Production and disappearance rates of fuel combustion products
are calculable as functions of altitude. Table 10 displays the principal con-
stituents of the "inviscid core" of a mature exhaust plume corresponding to
the fuel composition of Table 9o
It is important to realize that this simulation is for a mature plume
calculated without regard to entrainment and reaction of ambient air.
"Staging" occurs at 140,000 ft. Prior to this, a variety of transient free
radical and stable species have been produced and destroyed at plume temp-
eratures < 2700 K. Entrained air can react with or transport some of these.
Species present (and mole fractions) at 2690 K and 8.2 atmosphere were
calculated to be (Ref. 58):
AIG* (.0037), AtCl2 (.0023), AIC*3 (.0005)
AlOCl (.0019), AIOH (.0003), A^O2H (.0005)
CO (.252), CO2 (.021), Cl (.0038), FeC^2 (.0013)
H (.010), HCl (.154), H2 (.293), H2O (.163),
NZ (.091), OH (.0015), and minor products
About 99% of the air in the exhaust plume is entrained air, indicating
that the effect of reactions with air cannot be ignored. Predictive modeling
now indicates that two -phase flow, afterburning, and radiative energy trans-
fer combine to increase the calculated heat content of the plume more than
three-fold over pre-1973 estimates. Consequently most cloud-plume and
dispersion calculations previously reported are invalid. Plume rise is very
much greater than formerly supposed. Ground -level HCjf concentrations are
therefore lower, but very much more HC? is injected directly into the upper
troposphere (Ref. 57).
Particulate matter is created directly in solid propellant combustion.
It is possible also that free radical reactions can lead to formation of addi
tional condensible organics. One must ask if condensation nuclei are being
created. This concern must be related to the formation of enormous quanti-
ties of water vapor. Will precipitation result? Would precipitation wash
out much of the particulate burden, HC4, and AjfC^? If so, would this re-
duce air pollutant levels at the expense of local acid rainfall injury? Will
triboelectric charge generation have detectable chemical and meteorological
effects? Meteorological and cloud physics predictive modeling are being
pursued at Marshall Space Flight Center (Ref. 57) in an attempt to answer
50
-------
TABLE 9. TYPICAL SOLID FUEL COMPOSITION FOR SPACE
SHUTTLE LAUNCH VEHICLE (REF. 58)
%
16.0
69.6
12.0
0.4
2.0
Material
Reducing agent
Oxidizer
Binder
Binder
Stabilizer
Composition
Metallic AIL
NH4 CK)4
C6.9H10.1°0.3N0.3
(equiv. formula)
Fe2°3
C6.2H7.0°1.2N0.03
(equiv. formula)
TABLE 10.
CALCULATED COMPOSITION OF THE INVISCID CORE
OF A TYPICAL SSLV EXHAUST PLUME
(Fuel Composition, Table 9, p = 8 x 10"4,
atmos, T=471K)(Ref. 58)
Mole
Mole
Constituent Fraction x 10 Constituent Fraction x 10"
H2
co2
HC.0
N2
CO
470
200
150
94
67
AIC*3
C2H43
H2°
NH3
FeC£,
9.6
6.6
5.7
0.05
0.01
51
-------
such questions. Atmospheric monitoring of cloud plumes from Titan
launches is being performed by a large group from Langley Research Center.
Johnson Space Center personnel are responsible for ecological aspects of
SSLV launch studies.
The present state of the art is that predictive modeling results and
monitoring of plume behavior are now converging. It is becoming possible
to make conservative overestimates of exhaust plume effluents and ground-
level concentrations. Other areas of environmental concern (e.g., cloud
physics) are proving less tractable (Ref. 57).
Other contractors involved in dispersion, windfield, cloud physics,
or other environmental problem areas include System Analysis, Inc.,
Lawrence Liver more Laboratory, Brookhaven National Laboratory and
University of Pennsylvania. EPA may find it desirable to maintain close
liaison with the NASA facilities performing and directing plume composition
and dispersion studies, etc.
Nuclear Materials: It is proposed to launch nuclear cargoes for sub-
sequent fueling of nuclear propelled space tugs (Ref. 34). This idea should
be viewed with great wariness.
The consequences of a launch abort followed by atmospheric burnup
and/or disintegration are staggering to contemplate. Injection of nuclear
material into the earth's atmosphere or its distribution onto the ground and
into surface waters must be prevented. Perhaps it is possible to design
an absolutely impregnable containment which would not preclude subsequent
entry by space station crews. EPA may wish to maintain close liaison with
appropriate ERDA and NASA offices.
Microwave Radiation: Power generated in the proposed photovoltaic
panels is to be converted to 3.3 gHz microwave radiation and beamed to
earth receiving and rectifying antennas ("rectennas"). Williams (Ref. 34)
writes of beam intensities of 500 to 1000 W/m2 at the center of rectennas,
10 to 100 W/m at the edges, and an overall transmitting-receiving effic-
iency of 68%. He suggests control by phase-locking transmitter elements
onto a pilot signal originating in the rectenna center. One purpose of this
proposed arrangement is to preclude satellite attitude errors causing inad-
vertent high intensity microwave irradiation of populated areas. He believes
security fences would adequately restrict the general public to areas re-
ceiving less than 1 W/m . (1% of U.S. standards and about 10% of Eastern
European standards.) Further, he suggests that microwave intensities under
the rectennas would be less than 10 W/m , low enough to permit industrial
s itin g.
Microwave radiation is non-ionizing; its most apparent effect on tissue
is heating. Whole body 3.3 gHz irradiation at 100 W/m would dissipate no
more than 57.5 W in a human target, an amount said to be easily dissipated.
For perspective: the basal metabolic rate of a resting human is 70 to 90 W,
and about 290 W during moderate work. The U.S. standard of 100 W/m2 for
human whole body microwave irradiation is derived from these figures
(Ref. 59).
52
-------
"Microthermal" effects are nonuniform heating due to nonuniform
tissue absorption,, Some studies suggest adverse microthermal effects may
occur below 100 W/m2. Soviet researchers think abnormal nonthermal
effects occur also (Ref. 60); the USSR standard is therefore 0.1 W/m. This
is 1000 times more stringent than the U.S. standard.
It seems that great care must be exercised in developing specifications
for microwave power transmission target accuracy and securing safe zones
around rectennas. Regardless of whether or not somatic or genetic damage
could occur at 100 W/m2 irradiation, most humans would probably react
with anxiety to an unseen source of body heat approximating 70% of their
basal metabolic rate. The effects on aircraft passengers and birds need
careful assessment.
In addition, effects on radio transmission and the possibilty of inad-
vertent potentials induced in materials (e.g., blasting cap leads) need to be
assessed.
For a more thorough review of these points and others, Williams'
paper (Ref. 34) and Dickson's (Ref. 7) and their bibliographies should be
consulted.
Noise: The terrestrial receiving array should be quiet, except per-
haps for 60 cycle hum in peripheral equipment such as transmission lines.
The many SSLV launches (several per day cf. Table 6) may each approxi-
mate the noise generated by a Saturn V launch (Ref. 34). It seems unlikely
that these launch noises would be perceived forever by all residents near the
Kennedy Space Center as being desirable attractions. Landing shuttle orbi-
ters would create sonic booms. Perhaps resiting the program to a remote
location might be environmentally desirable. It is beyond the scope of this
report to analyze the economic viability of resiting.
Materials and Energy Requirements: Because designs are still con-
ceptual and changing, it is impossible to make precise materials and energy
estimates. Glaser (Ref. 61) stated that the energy required to fabricate,
launch and construct the satellite would correspond to this schedule:
Months of Operation
Materials to Pay Out
Propellants 6
Solar Cells 3
Ground Support 3
Facilities
Williams (Ref. 34) believes two years to be realistic. Dickson (Ref. 7) cal-
culates that Glaser1 s design would require 10 kg of Ga in GaAs cells —four
times more Ga than the U.S. Geological Survey anticipates will be the cum-
ulative total production by year 2000. Williams, loc cit, has correctly
pointed out that structural material requirements should be very much lower
than for terrestrial applications.
53
-------
Attention to materials demands is essential, and may profoundly in-
fluence design features. Commitment of a large fraction of GaAs produc-
tion capacity to geosynchronous satellite power stations would preclude its
use in large terrestrial solar applications.
Hydrogen Cycle Storage — Direct Effects
The hydrogen cycle (water electrolysis cell-hydrogen/air fuel cell) is
a particularly good match to solar electric generation because it requires dc
power, uses modular construction, and therefore can more rapidly track
load projections with shorter lead time and is applicable to grid dispersed
facilities. It is much more efficient than oil fired turbines, is not site
specific, is quiet and clean and does not consume significant quantities of
water (makeup water requirements are 5% or less) (Ref. 27),
Principal components of a hydrogen cycle system are electrolysis cells,
hydrogen storage equipment, fuel cells, power conditioning and handling
equipment, and other peripherals. In some systems the fuel cells and elec-
trolysis cells can be the same units. Power conditioning equipment possibly
may be common to the generating station. Hydrogen storage methods in-
clude storage as FeTiH2 (Ref. 51) using low grade heat for degassing. A
detailed unit operations oriented environmental assessment of a 500 MW fuel
cell peak shaving plant was performed by Sears (Ref. 21), Figure 15 is a
flow diagram for the plant, and Table 11 summarizes the environmental re-
leases. A significant problem is release of asbestos in electrolysis cell
off-gas and cell flush waters. Recent progress in development of solid
polymer electrolyte/separators may obviate the problem. Release of pyroly-
sis products of SPE (typically fluorosulfonated polymers) in the event of a
cell fire needs to be evaluated.
Alternative Subsystems
Steam Reforming: Complete reliance on electrolysis for H-> produc-
tion may be impossible, because peak shaving may be required in excess of
off-peak solar generating capacity. Optimum utilization of existing facilities
suggests steam reforming of distillate oil and oxidation of hydrogen in the
fuel cells. Steam reforming is the only applicable, well developed alternative
to electrolysis. Design, performance and atmospheric emissions are de-
scribed by Lueckel and Farris (Ref. 52). Based on performance of a 26
MW"e plant, emissions are:
Pollutant lb/106 Btu Input
NOV 0.020
X c
SO2 3 x 10
Particulates 3 x 10~6
Smoke Nil
Noise Suitable for Residential
Area
54
-------
Ul
tn
Electrol ysis
Modul e:
1 ^
H2
i
Dryer
Potable Water Treatment
Sanitary Facilities
Laboratory Facility
Heating Plant
Shops
Emergency Generatore
Misc. Waste
Water
Streams
^ Recirculating
Cooling
Towe r
V. Process Water
\
1
J
Sin
~+^
\
o>
i»
\ \
o.,V "* \
a: % 9-4> Q*—_
'A off- T 0 o \ , _,
FB'dr HXH-
n
H 0
~\ ucnt G i Ltc |
Peak j ^,-a e*K
I /^
s__ Inverter
llUlllljlUWlls ^UllVtlLtl
Ferry Nuclear F.icilityl | eirld
1 1
1 1
1 |
t L'l c"""°' >J J
__.-,'1 5
_c
'Jl
T
C
we r ^
— « — * ^
V La^o.j.
--«$ To^
OJ
Tj
^
TStTl Ttr_
^ ^-IX}*-
Lj,_j
ve r Blow-dow
Clarifying
and
Dum i n c ral -
Facility
So *
K
"n f
r-
Figure 15. Unit operations detail for an electrolysis —fuel cell hydrogen cycle (Ref. 27).
-------
TABLE 11. SIGNIFICANT ANTICIPATED DIRECT
ENVIRONMENTAL RELEASES FROM
A 500 MW ELECTROLYSIS-FUEL CELL
HYDROGEN CYCLE FACILITY (REF. 27)
Component
Process Water
T rpatment
Fa> ility
Elcctroysis
Units
Hydrogen
Sto rage
Fuel Cells
Cooling Towers
Power
Handling
Section
Sanitary
Facility
Atmospheric
Emis sions
None.
Option (a)fAsbestos
Option (b): Deteriorated
SPE Entrained in
65,600 Scfm Op.
None.
Excess Air
Fog and Icing, Drift
(containing Treatment
Chemicals ) .
Heat.
Noi s e.
Traces PCBs (SF6?)
from Transformers.
Herbicides from
Rights-of Way.
Smoke from Slash
Burning.
None.
Waste Water
Discharges
Filter Bed Backflush
(to River).
Demineralizer Regen-
ration (to Lagoon).
Option (a):Asbestos in
flush (to Lagoon).
Option (b): Deteriorated
SPE in Flush (to Lagoon)
None.
Option (a): Asbestos in
Product Water and
Flush.
Option (b): Deteriorated
SPE in Product Water
and Flush.
Treatment Chemicals in
Slowdown (to Lagoon).
None.
Pathogens, BOD,
Treatment Chemicals
Solid
Waste
Occasional
disca rd of
s po iled resin
(to landfill).
Sludge
(Cf. Waste
Water).
Discard of
Spoiled Fill (to
oxide conver-
sion or recycle
facilities) .
Sludge
(Cf. Waste
Water)
Sludge
(Cf. Waste
Water).
None.
Sludge
Accident or
Malfunction
Malodo rous
Emissions from
Anaerobic
Processes in
Filter Bed
Pyrolysis
Products of
SPE
FeTi and FeTiH^
Granular and
Powdered
Material (to work
pi a eel .
Pyrolysis
Products of
SPE.
Wood Smoke and
Pyrolysis Pro-
ducts if Timber
Construction
Used.PVC Pyroiy-
sis Products
PCB's in
Explosion
(SF6).
Raw
Sewage
56
-------
Cryogenic Storage of Oxygen and H2
Both technologies are very well developed due to NASA R&D. They
require energy expenditure to perform liquefaction and maintain vacuum.
Liquid hydrogen is much less safe to store and handle than FeTiFU.
H2 —oxygen fuel cells, using alkaline electrolyte, are better developed
than H2 —air fuel cells, which must use acid electrolyte (Ref. 53). Periodic
flushing of the fuel cells will require different waste treatment strategies in
the two cases.
Hydrogen as a Fuel
Another proposal (Refs. 28, 54, 55,for example) is to produce hydrogen
by electrolysis, but to transport it to demand centers as pipeline gas and
combust it on site in conventional gas fired applications. No fuel cells would
be required, but pipeline transmission corridors would be required. Inevit-
ably, safety considerations would need detailed study. Hydrogen combustion
(pure or CH4-blend) should be particularly clean, but NO emissions would
require evaluation for specific burner designs.
Resource Commitment and Depletion
Silicon: Total domestic and world reserves are huge. Published data
(Ref. 62) suggest that several hundred million tons of 95% SiC>2 are available
domestically. Known reserves of 98% SiO-> are sufficient to support 10^ tons
(9.1 x 10*0 kg) Si production. Few states have published survey data; there-
fore, the true reserves exceed the known reserves by an uncertain, but
certainly large margin. Figure 16 (Ref. 62) presents production and demand
projections without regard to major solar power applications. Domestic pro-
duction is only one-third of world production.
All silica mining, incidentially, is open-pit, with a small over-burden
to ore ratio.
Comparison of Tvend Projections and Forecasts
(or Primary Silicon Production.
Comparison of Trend Projections and Forecasts
for Silicon Demand.
Figure 16. U.S. silicon production and demand projections, without
regard for major solar power applications (Ref. 62).
57
-------
Current Si solar cell production is a small fraction of total Si use.
Ln 1975, for the first time, terrestrial applications exceeded^pace appli-
cations (1000 m2 versus 500 m2) (Ref. 17). Assume 40 W/m capacity,
(24 hr average in favorable sites),and 175 /im thickness. This is intermediate
between the 250 jLLm current devices and Prince's, goal of 100 jam devices
(Ref. 6). A 1 GWg facility would require 25 x 10° m2 of panel with a volume
of 4375 m3. This is 10.2 x 10° kg or 11,300 short tons.
Conclusion: 50 GWe silicon photovoltaic capacity in
the U.S. in year 2000 would increase forecast de-
mand 38 to 58% but would have minor impact on
domestic reserves.,
Cadmium: This element is fairly rare, although it sees numerous in-
dustrial applications. Figure 17 depicts supply/demand relationships for
Cd which must be considered in any massive solar power development using
CdS. The U.S. already uses about one-half of world production.
Cadmium is almost wholly a byproduct of zinc production. Its avail-
ability is tied to zinc demand, because it does not produce sufficient income
::,inc producers to adjust production schedules to meet Cd demand.
Apparent recoverable domestic cadmium reserves are estimated to
be 8.6 x 10 kg. World reserves including U.S. are estimated to be 6.4 x
10^ kg. Recent discoveries in middle Tennessee may increase domestic
reserves by only 5 to 15% (Ref. 62). About 25% of our cadmium is imported
(Ref. 63). Demand-production trends are given in Figure 18.
Using Project Independence numbers (Ref. 23), viz 5.0 x 10 kg Cd/
GWe, we find that the estimated total recoverable domestic cadmium re-
serves correspond to only 17 GWe of generating capacity. Actually, the
Project Independence numbers were derived from 20 jLtm CdS in CdS/Cu2S
panels, with an assumed efficiency of 7%, Substantial reductions in film
thickness and perhaps a small improvement in efficiency can be anticipated.
Nevertheless, CdS appears to be resource limited.
The situation may be more serious than for gallium, because unlike
the latter, cadmium sees widespread industrial application. Commitment
of cadmium to solar development may be expected to force other users to
seek substitutes which may be environmentally less desirable. An excep-
tion is electroplating. There is already a tendency to substitute zinc for
cadmium.
The cadmium supply/demand situations created by solar power develop
ment therefore seems to merit attention by EPA, ERDA and various U.S.
mineral resource agencies.
Gallium (Refs. 62, 64,65): Although gallium is relatively abundant,
there do not appear to be any deposits of gallium ore rich enough to justify
processing for gallium alone. The apparent exceptions are germanite which
may contain 1% Ga, and gallite, CuGaS2 found in South Africa. However
they are too rare to be counted as resources. Many coals which contain
58
-------
WORLD PRODUCTION
26,460
1
1
West Germany
e 400
Poland
«800
U.S.S.R.
e 3,000
1
United States
e 3,780
Mexico
6 3,580
Canada
e6,400
Peru
e2,100
Congo
(Kinshasa)
MOO
Japan
e 1,100
Australia
e 1,700
Other
e 3,300
1 605
2,5
(A)
50 >
00
450
20 i
346
(B)
152
11 ^
_661>
007
- — *
r — "
U.S. refinery
production
10,651
Imports, metal
1,927
Industry stocks
1/1/68
1,541
Stockpile
release
808
b
U.S. supply
14,927
fe
Industry stocks
12/31/68
1,069
Exports
530
U.S. demand
13,328
KEY
e Estimate
(A) Contained in imported flue dusts
(B) Contained in imported zinc concentr
SIC Standard Industrial Classification
Unit: Thousand pounds of cadmium (Cd)
jtes
b
It
fe
*
^
. ^
Motor vehicle parts
(electroplating)
f5/C 37 U)
e 1,300
Aircraft & boats
(electroplating)
(SIC 37? & 373)
C3CO
Electroplating,
other ^
(S/C 3-171)
e 5,900
Primary batteries
(SIC 3692)
e 400
Pigment, inorganic
(SIC 2818)
e 1,500
Plastic material?
{SIC 2821)
e2,500
Other
«928
Government stockpile balance - 12,940
Figure 17. 1968 supply-demand relationships for cadmium (Ref. 62)
(units in 1000 Ib or 454 kg).
-------
S JO
10
Detnonti
Z4O
21.2
6.0
EOOO
Figure 18. Comparison of U.S. trend projections and forecasts
for primary cadmium (Ref. 62).
germanium also contain gallium. Fly ashes of these coals occasionally con-
tain as much as 0.1% Ga. This cannot be considered a dependable resource
upon which to base a solar power industry. Basically, gallium is a by-
product of aluminum and zinc refining.
Uncertain data suggest domestic reserves are 2 x 106 kg in bauxite,
and 0.7 x 10b kg in zinc blende (Ref. 62), About 96% of our bauxite is im-'
ported, however (Ref. 63). Gallium supply problems may encourage some
recycling of retired GaAs devices, reducing long term (but not short term)
demand as well as reducing environmental releases of Ga and As. Figure
19 presents production and demand forecasts for Ga without regard to major
solar developments.
Assuming 1 ^tm thick GaAs (Ref. 7) generating about 20 W/m2, a 1 GW
plant requires 50 x 10b rr/, 50 m3, 2.67 x 105_kg*", or 294 short tons. This
corresponds to 141 short tons of Ga (1.28 x 105 kg), about ten times cumula-
tive domestic production 1975-2000, assuming 500 kg/yr.
I*
Using density calculated from crystal structure assuming cubic cell
a0 =5.646, Z =4 (Ref. 66).
60
-------
flOO
1,200
1.000-
1,150
•-690
194
1968
2000
Comparison of Trend Projections and Forecasts
for 1'rimary Gallium Production.
1949
1968
2000
Comparison of Trend Projections and Forecasts
for Primary Gallium Demand.
Figure 19. U. S. gallium production and demand projections,
without regard for major solar power applications
(Ret 62).
If U.S. reserves are indeed only 207 x 10 kg, the requirements of one
1 GW plant would represent nearly, 5% of our domestic reserves. World
reserves are estimated at 110 x 10 kg (Ref. 64).
The Project Independence data (Ref. 23) create an even bleaker picture.
They assume 1.4 x 10° kg/GWe —about 112 times cumulative domestic pro-
duction of 500 kg/yr 1975-2000. Our experience with dependence on foreign
oil suggests ERDA may wish to monitor GaAs utilization and supplies with
some care.
Recently Varian announced develop'ment of GaAs capable of withstand-
ing the higher temperatures needed in concentrators. Should this develop-
ment prove to be practicable in large facilities, GaAs photovoltaic appli-
cations would cease to be resource limited.
Arsenic: The U.S. does not seem to face resource limitations with
this element. U.S., domestic reserves are said to be 1.7 x 10° kg and world
reserves 3.8 x 10' kg (Ref. 62). A 1 GWe, 1 /imGaAs facility would require
1.4 x 10 kg or 152 short tons. That represents about 50 years' domestic
production at current levels (Figure 20) and about six times current domestic
annual demands but only 0.008% of U.S. reserves. GaAs applications will
be limited by Ga availability, not As.
Miscellaneous Candidate Photovoltaic Materials: From time to time
other materials are suggested and some of these may achieve substantial
61
-------
1968
2000
Figure 20. U.S. arsenic demand and production trends, disregarding
solar development (Ref. 62).
62
-------
use. For this reason, we supply Table 12 which presents some data needed
to judge resource commitment. We include GaAs, CdS and Si for comparison.
This table does not show how much of the indicated reserves are com-
mitted by other uses at current or foreseen application rates. In some
cases, current uses exceed known reserves as in the case of copper. It
does not indicate improvements in reserves possibly developed by improved
byproduct recovery methods, new mineral explorations, etc. Most import-
antly, some of these reserve data are based on current price structures.
If prices increase dramatically, former "nonrecoverables" in low grade
ores or base metals will suddenly become "reserves," as happened to
mercury in the 1960s.
Entirely new supplies of trace metals, and even sulfur, can develop
from pollution control technology. Sulfur from fuel and flue gas desulfuriza-
tion, gallium from flyash, and selenium and tellurium from smelter waste
stream recovery are examples. Detailed discussion of these subjects is
entirely outside the scope of this report. Nevertheless, these matters may
assume overwhelming importance and appropriate offices of EPA, ERDA
and the Department of the Interior, for example, may wish to keep each
other informed.
Glass and Aluminum: Current and anticipated technology require
hermetic seals on the front surface of thin film (CdS and GaAs) photovoltaic
panels. Research is underway to develop CdS panels with the semiconductor
sprayed directly onto float glass. In any event, some sort of long-lived rear
support surface will be required, which may be the conductor.
Suppose each photovoltaic panel requires a 3/32 inch glass front
surface and an 1/8 inch aluminum back surface. Assuming 50 x 10° m /
GWe (538 x 106 ft2/GWe) we need for each GWe capacity:
4000 x 106 kg (4.4 x 106 tons) glass
5100 x 10 kg (5.6 x 10 tons) aluminum
This aluminum requirement will be in addition to any used in support
structures, transmission towers, and transmission lines.
Lead: This element would find potential application in storage devices.
Although lead would probably be unsatisfactory in the long terra (cf. the
previous discussion) it will possibly see application in the earlier pilot and
demonstration plants. We recommend that a review of technology and re-
source demands of lead-acid storage cells of utility size be performed.
Extrapolation of lead requirements from automotive and electric powered
vehicle use is likely to be an invalid exercise.
Domestic and world reserves of lead are believed to be 3. 2 x 10 and
5.4 x 10 10 kg (35.3 x 106 and 60 x 10° short tons), respectively (Ref. 62).
Lithium (Refs. 62, 67): Experimental Li, A^-iron sulfide cells have
been demonstrated with a storage capacity of 150 W-hr/kg (compared to
63
-------
TABLE 12. SUMMARY OF SOME PHOTOVOLTAIC MATERIALS DATA
01
Material
Si
CdS
Cu2S
GaAs
CdTe
CuInSe7
L*
InP
Density
g/cm2
2.33
4.82
5.6
*
5.34
6.20
5.7
4.79+
Mol.
Wt.
28.09
144.46
159.14
144.64
240.00
336.28
145.79
Element
Si
Cd
S
Cu
S
Ga
As
Cd
Te
Cu
In
Se
In
P
Mass
Element in
1 m2 Panel
1 11 Thick, g
2.33
3.76
1.06
4.5
1.1
2.56
2.78
2.91
3.29
1.1
1.9
2.7
3.78
1.01
Assumed Thick-
ness, Id in
Application
175
20 (Ref. 23)
1 (Ref. 23)
1 (Ref. 7)
1 (Ref. 23)
Unk.
Unk.
Mass
Needed in
1 GWe
Plant,
106
25
5
N/A
0.14 (Ref. 23)
N/A
0.27
0.14
2.4 (Ref. 23)
0.7 (Ref. 23)
Unk.
Unk.
Unk.
Unk.
Unk.
Reserves
106
U.S.
Unlim.
86
3x 105
77,100
3x 105
2.7
1723
86
-,##
77,100
0.4
23
0.4
Adeq.
(Ref. 62)
kg
World
Unlim .
635
2x 106
Unk.
2x 106
110
3810
635
63**
Unk.
2.3
108
2.3
Unk.
#*
Calculated from crystal structures using a = 5.646A, Z = 4 (Ref. 66).
Improved byproduct recovery practices could increase this (theoretical limit <_2x).
Calculated from crystal structure using a = 5.869A, Z = 4 (Ref. 66).
-------
25 W-hr/kg for lead-acid batteries). Thus, Li based batteries show un-
usual promise as light, high energy density storage components for electric
vehicles and utility applications* Demand projected to year 2000 for these
two applications is 2.7 x 10^ kg and 5.4 x 10 kg of lithium, respectively.
Additionally, fusion power reactors using lithium as fuel source, blanket
and/or coolant may need an estimated 100 to 1000 kg Li/MWe capacity. These
uses of lithium may not materialize. Alternative battery materials (e.g.,
sodium) may be employed instead, and fusion may not work.
Production of this metal during 1974 was 3.4 x 10 kg domestically and
1.6 x 10 kg in the remainder of the world. Past and projected demand and
production are displayed in Figure 21 in which fusion power and electric
storage applications are disregarded. Production for 1974 fits almost
exactly onto the 20 year trend line, rather than into the constant-ratio fore-
cast region.
1968
Comparison of Trend Projections and Forecasts
for Lithium Demand.
1968
2000
Compaiison of Trend Projections and Forecasts
for Lithium Production.
Figure 21. U.S. lithium demand and production projections disregarding
solar and fusion power application (Ref. 62).
New discoveries of lithium reserves and expanded exploitation of lower
grade deposits make the reserve figures of Ref. 62 rather meaningless now.
At a January 1976 Conference on U.S. Lithium Resources and Requirements,
rather disparate points of view were expressed on the adequacy of lithium
for the new storage and fusion applications (Ref. 67). Vine of USGS thinks
65
-------
domestic reserves are less than 109 kg, barely more than the projected
storage battery requirements. Kunaoz of Foote Mineral Company thinks ore
reserves are sufficiently large —and elastic —to satisfy conventional and
new energy needs.
There is no concern about reserves being able to support conventional
applications: the Kings Mountain area of North Carolina has at least a 100-
year supply.
In assessing indirect effects of resource commitment, one must con-
sider whether dedication of a major share of reserves or productive capacity
will force other users to exploit alternative materials much less desirable
environmentally. Clearly this can be an open-ended question, and to per-
form such a study is beyond the scope of this report. It needs to be done,
however.
In the particular example of lithium, we could face an even more
disastrous indirect effect. For if fusion power is developed successfully,
we may have to dedicate much of the world's recoverable lithium to energy
production, proscribing its use in other major uses, such as storage systems.
Fortunately, utilization in Li, Ajf electrodes does not seem to be irreversible.
Developing methods of husbanding lithium, and recovering it from electrodes
and battery electrolyte would seem to be prudent.
Indirect Effects
Energy to Produce Materials: Unknown quantities of aluminum, steel
and copper will be required. Recent reports by Battelle Columbus Labora-
tories (Ref. 68) state the unit energy costs (Table 13) of some of the materi-
als needed for solar power development. Their numbers include the entire
production cycle from mine face to product.
TABLE 13. ENERGY COSTS FOR SELECTED PRODUCTS (REF. 68)
Product Energy Cost
(kWhr./kg)
Aluminum Ingot 79.0
Portland Cement 2.5
Refined Copper 36.0
Glass Containers 5.6
Carbon Steel Castings 13.6
Refined Lead g.7
Elemental Phosphorus 55.0
Sulfuric Acid 0.01
Zinc, Elemental 21.0
66
-------
Certain of the photovoltaic semiconductors are byproducts of copper,
aluminum, or zinc manufacture. However, the additional energy costs be-
yond Cu, Aft or Zn production required to produce Cd, Ga, As, etc., are
not readily available..
Project Independence (Ref, 23) estimates ZOO tons steel/MW peak, or
2.47 x 10° MWhr/GWpk capacity to produce the steel alone. This would have
to be multiplied by a capacity factor, depending on duty and installed per-
formance.
Taking our previous estimates for glass and aluminum in the panels,
assuming flat glass and glass containers have similar energy costs, and
ignoring aluminum rolling mill energy costs, we have
4000 x 106 kg glass/GW
equivalent to 2.2 x 10? MWhr/GW
5100 x 10 kg aluminum/GWe
equivalent to 4 x 108 MWhr/GW
These numbers correspond to 3/32 in0 single glazing and 1/8 in. backside
aluminum support only. Other uses of glass would be minimal unless CdS
deposited on float glass prevails. Assuming energy costs are reflected in
dollar costs, utility companies can decide on a rational basis whether it is
cheaper to use glass backside support with thin film conductor, or aluminum.
Estimates of the aluminum needed for transmission line conductors and
towers can be made easily when siting is decided.
Lead for storage battery use (a dubious prospect) is estimated thus:
suppose 60% of battery mass is lead, and storage capacity is 25 Whr/kg
total. Then we find 207 MWhr. energy cost/MWhr. storage capacity, using
Battelle's numbers. Sulfuric acid energy cost is trivial.
The energy cost for silicon has been determined by Battelle but was
not quoted in Ref. 68.
Phosphorus energy costs might be non-trivial if InP should find
application, but the data in Table 12 are inadequate to compute this.
Air Pollution to Produce Materials: To do this properly, one would
combine projections of various types of solar electric power facility con-
struction, projections of coal, oil, and gas fueled steam electric power
generation with appropriate mixes of fuel characteristics, compliance
schedules, changing primary standards and state implementation plans, etc.
There are too many uncertainties now to make this a fruitful effort.
67
-------
Let us assume (incorrectly) that all power needed to manufacture
materials is produced in 32% efficient, pulverized bituminous, drybottom
utility boilers. Further, let us assume 3% sulfur, 12% ash, 12,800 Btu/lb
coal.
Reliable emission factors (Ref. 69) are available. Energy production
would be 2,4 MWhr/ton of coal. Emissions before and "after" control are
given below in Table 14, We assumed 90% efficient precipitator s, and time-
linear increase from zero to 100% flue gas desulfurization (time average
~50% control of SOJ.
TABLE 14. EMISSIONS (lbs/MW hr.)
Item
No Control Time Average
Control
Particulates
SO
X
NO
X
85.0
48.0
7.5
8.5
24.0
7.5
Using previously discussed energy costs, we find the air pollutant produc
tion presented in Table 15.
TABLE 15. SOME AIR POLLUTANTS PRODUCED DUE TO ENERGY
COSTS OF MATERIALS PRODUCTION FOR SOLAR
ELECTRIC FACILITIES, PER GWe
INSTALLED CAPACITY
Particulates SOX
Commodity (106 Ib) (106 Ib)
Carbon Steel
Glass Glazing
Aluminum Back Panels
Totals
21
187
3400
3608
59
528
9600
10,187
NOX
(106 Ib)
19
165
3000
3184
Primary aluminum production will result in emissions of particulates
(including particulate fluorides) and gaseous fluorides. Fluorides cause
vegetation and livestock injury.
Emission factors for many unit operations and for a number of control
options are available (Ref. 69). The combinations are too numerous to
attempt application here. Generally, total fluorides emissions run 15 to 30
Ib/ton uncontrolled and 0.02 to 5 Ib/ton for various control options, for
68
-------
each of several unit operations. We anticipate fluoride emissions to be 0.3
x 106 to 30 x 10b lb/GWe.
Similar uncertainties apply to direct emissions from steel manufacture
(Ref. 69) including particulates, carbon monoxide, and fluorides. For
example, CO emissions may run 3,6 x 10 lb/GWe capacity if electric arc
steel is used, but negligible in the case of BOF production.
Glass melting produces particulates and fluorides. Perhaps 9 x 10 Ib
particules/GW capacity would be an upper limit, since control strategy
should improve. Fluoride emissions may be less than 90 x 10 Ib/GW ,
assuming improved control strategy.
Direct emissions of cadmium during smelting, refining, sulfide manu-
facture, purification, solar cell manufacture, etc., are all significant
problem areas. Fleischer et al (Ref. 70) have performed an extensive re-
view for the NIH Panel on Hazardous Trace Substances. Photovoltaic
material manufacture was not considered, although base metal smelting was,
since almost all cadmium is a byproduct of zinc, copper and lead smelting.
The data suggest that probably nearly all airborne cadmium is due to
man's activities. Only inadequate data are available on the fate of air-
borne cadmium and its residence time in the atmosphere. Urban areas are
found to have concentrations from 100 to 400 ng/m-^. The sources are
thought to be primarily metallurgical operations. Table 16 presents esti
mates based on production data and assumptions on losses and fates of
waste streams. The work is by Davis et al as quoted in Ref. 70. No
measurements were involved and some of the input assumptions have been
questioned (Ref. 70)c
Tables 17 through 20 present some analytical data from flue dust
precipitator dusts and other waste streams in zinc beneficiation operations.
Clearly a large increase in atmospheric cadmium levels will accompany in-
creases in zinc production regardless of photovoltaic needs unless zinc
smelter emissions are controlled more effectively. A mixed blessing is
that the major fraction of zinc smelter cadmium emissions accumulates in
the surrounding soil, where it penetrates deeply (Ref. 70). Figure 22 pre-
sents data of Miesch and Huffman (as quoted in Ref. 70) for soils near a
smelter which had been operating for 80 years. Naturally this creates a
potential for surface water and ground water contamination, although little
is known about the fate of Cd in the hydrologic cycle. It also creates a
potential for contamination of vegetation used as foodstuff by man and by
livestock. For further details, the reader is urged to review the report by
Fleischer et al (Ref. 70) since it is the most comprehensive review of
environmental cadmium in existence.
Particularly if CdS photovoltaic material should become a high demand
commodity, we can expect the price to be driven up. If this should happen,
we might anticipate some smelters would adjust their operations to improve
by-product recovery of cadmium, to the benefit of atmospheric quality.
69
-------
TABLE 16. ESTIMATES OF ATMOSPHERIC EMISSIONS OF CADMIUM
IN THE U.S. FOR 1968 (REF. 70)
Item
Cd, Ib x 10
Processing:
Mining and ore processing
Smelting
Reprocessing of metals
Electroplating
Pigment manufacture
Plastic manufacture
Alloys
Battery manufacture
Miscellaneous
Consumptive uses:
Wear of automobile tires
Burning of oil, motor vehicles
Fungicide use
Fertilizers (superphosphate)
Steel scrap reclamation
Radiator scrap reclamation
Plastic and pigment incineration
0.53
2100,00
33.53
Very low
21.00
6.00
5.00
0.40
1.13
11.40
1.82
0.50
0.91
2000.00
250.00
190.00
4622.22
(= 2300 tons)
Until a better unit operations analysis and material balance data for
smelting and for CdS production are available, quantitation of atmospheric
emissions seems hopeless. Davis' results (Table 16) cannot be scaled
validly. For example, his estimate for smelting emissions represented
about 20% of current production (Ref. 70).
We have no comparable study of gallium. Such a project would be
valuable, especially if it appeared that gallium arsenide would find massive
application in photovoltaic generating stations (or geosynchronous satellite
microwave transmission). Not one reference to gallium was found in the
pollution abstract literature for the period 1973-75, for example.
70
-------
TABLE 17. COMPOSITION OF ORE CONCENTRATE
AND REPRESENTATIVE SAMPLES OF FLUE
DUSTS FROM ROASTING AND
SINTERING OPERATIONS (REF, 70)
Sample
Ore concentrate
Flue dust from roasting1 furnace
Dust from electrostatic precipita-
tors from roasting furnace
Flue dust from sintering
machine
Dust from sintering machine
collected in cyclone
Dust in exhaust from cyclone
Agglomerate
Zn/Cd
256
246
48
98
136
5
356
Cd, %
0.18-0.20
0.1-0.18
0.6-0.66
0.55
0.27-0.55
3.6-8.8
0.16
Zn, %
45.9-52.6
30.5-38.9
22-33
54
53.4-58.1
22-44
60.1
Pb, %
1.8-2.9
2.5-5.0
20-34
2.1
1.6-3.1
95-37.1
1.6
Fe, %
6.7-10.8
4.8-6.3
3.1
—
—
—
9.6
Cu, %
0.6-1.6
0.7-1.0
0.35
1.2
1.1
0.22
1.28
S, %
25.6-32.2
15-19
—16
3.0
4.3-5.5
7.1
1.37
TABLE 18. COMPOSITION OF ZINC AND CADMIUM
COMPOUNDS IN THE DUST FROM AN
ELECTROSTATIC PRECIPITATOR OF A
ROASTING PLANT AS DETERMINED BY
SOLVENT EXTRACTION (REF. 70)
Solvent
Water
3% HtSO»
10% HiSO.
Residue
Extracted
Cd
Zn
Compounds
extracted
69.0
8 fi
14.5
67.0
73
14.0
6.7 13.9
Sulfates
"Pj-p<» nvi
Ferrites
ZnOFe.O.)
Sulfides
71
-------
TABLE 19. DISTRIBUTION OF CADMIUM IN PRODUCT STREAMS
FROM COPPER SMELTERS OF DIFFERENT DESIGNS (REF. 70)
Cadmium recovery, %
Reverbatory Blast furnace
Stream smelting smelting
Converter matte 49.1-56.3 38.4
Dust collected 6.8-9.2 22.9
Waste slag 7.1-12.6 16.6
Gas losses 29.1-29.8 22.1
TABLE 20. DISTRIBUTION OF CADMIUM IN BESSEMER
PROCESSING OF MATTE FROM REVERBATORY
AND BLAST FURNACE SMELTING (REF. 70)
Stream Cadmium recovery, %
Crime copper —
Converter al&g 85.8
Flue dust 6.2
Gas losses 58.6
72
-------
IOOO
Log Cd-1.4226-1.3571 log D
'2/3 I
1-1/2 2-1/4 3-3/8 5-1/16 7-1/2 11
Distance from stack, D (mites)
17
Cadmium contents of soil samples at
depths 0—4 in. as a function of distance from
smelter stack. Data of Miesch and Huffman
ICO
10
£ '
a
a
O.I
'2/3 I 1-1/2 2-1/4 3-3/8 5-1/16 7-1/2 11
Distance rrornsiocK.Dimiies;
17
Cadmium contents of soil samples at
depths 6-10 in. as a function of distance from
smelter stack. Data of Miesch and Huffman
Figure 22. Soil contamination by airborne cadmium (Ref. 70).
73
-------
Arsenic exists in atmospheric particulates primarily in the form of
inorganic oxides and arsenates (Ref.71). Atmospheric As2O, concentrations
as high as 1.7 /Ig/m^ have been measured ? km from a copper smelter
(Ref. 72).
Literature on atmospheric arsenic exists because of interest by EPA,
Public Health Service and State authorities' interest in copper smelting, for
example. Clearly- it would be advantageous to EPA and ERDA were a careful
review of atmospheric arsenic emissions performed with emphasis on emis-
sion factors for unit operations, from smelting to photovoltaic panel fabrica-
tion.
An expanding body of literature exists concerning distribution of trace
metals as functions of particle size, and even radial distribution within
particles (Ref. 73). Arsenic and cadmium are found preferentially in fine
particles in the respirable range. This is a particle size range which is
particularly difficult to control. Natusch (Ref. 73) and others postulate that
a volatilization-adsorption mechanism may be responsible for the prefer-
ential distribution of toxic heavy metals in finer particulates. This is a
particularly unpleasant suggestion because high temperature processes are
..a; \r to smelting and beneficiation, and because we have evidence of prefer-
tiiLial Cd volatilization (Ref. 70) in zinc smelting.
Natusch (loc cit) suggests that a volatilization-adsorption process could
be exploited to develop a control system. He proposes preferential adsorp-
tion of volatile heavy metal species onto large, easily collectable heat stable
particles deliberately introduced into flue gas streams.
If electronic grade cadmium sulfide and gallium arsenide become major
commodities, it will certainly be necessary to investigate improved control
technology, especially in the fine particle range. Both EPAand OSHA may
find it useful to maintain close surveillance of the emissions and control
technologies in these new industries.
Methods of sampling and analysis of toxic trace metals in atmospheric
particulates are well developed and improving (Refs.71, 74, 75). Gallium,
however, has been overlooked. Many methods can be developed by elabora-
tion of water analysis methods (Ref. 76) using bubblers containing appropriate
absorbing reagents.
Air Pollution Prevented by Not Burning Fuels; It may be invalid to
assume that any air pollution will be prevented by solar power development
in the intermediate term. Even the most optimistic projections in ERDAs
scenarios (Ref. 1) predict solar electric's contribution to be less than 2% of
that of coal.
However, it is straightforward to calculate the emissions which would
have occurred were no solar electric capacity developed, and fossil fuel com-
bustion filled the gap. Assume 3% sulfur, 12% ash, 12,800 Btu/lb coal burned
in 32% efficient dry bottom pulverized coal utility boilers. Assume 90% effi-
cient precipitators (Ref. 69) in some distant time when the electric utility
74
-------
industry has achieved system-wide 90% flue gas desulfurization. The conse-
quences are given in Table 21.
TABLE 21. HYPOTHETICAL "POLLUTANTS PREVENTED" BY SOLAR
ELECTRIC SUBSTITUTION FOR COAL FIRED UTILITY
GENERATION (See text for fuel parameters)
Emissions Prevented (Ib/GWhr)
Pollutant
Particulates
SO
X
NO
X
No Control
85,000
48,000
7,500
Control
8,500
4,800
7,500
These numbers need to be multiplied by solar system capacity, capacity
factors, etc.
It is probably fair to assume that by the year 2000, for example, Fed-
eral and State stationary source regulations, as well as control technology
will be very different.
WaterPollution to Produce Materials: This subject has three major
aspects; (1) surface water degradation and groundwater disturbance due to
coal, zinc, copper and bauxite mining; (2) water pollution related to ore
roasting, smelting and refining, and (3) waste water discharges in CdS and
GaAs production and device fabrication.
Most bauxite is imported. Further, solar development is not likely to
enormously increase aluminum production. Copper and zinc mining is out-
side the scope of this report, and it is not clear that increased Cd and Ga de-
mands will necessarily increase aluminum, copper and zinc mining. Coal
mining will be treated in a later section.
Fleischer (Ref. 70) indicates that the behavior and fate of cadmium in
the hydrologic cy^le is largely unknown. A large but still inadequate litera-
ture on the aquatic ecology of cadmium exists. The current literature sug-
gests that there is no interest at all in Ga and In, and surprisingly little in
As.
Introduction of cadmium into surface waters due to mining and smelting
is significant. In addition, great quantities of material are disposed of in
slag heaps, dross disposal, dumps, etc. Ultimately, some fraction of this is
leached into surface and ground water. Table 22 and Fig. 23 due to Fleischer
et al (Ref. 70) present one estimate of the problem.
Much more remains to be done in this area. Fortunately, solar elec-
tric development will be slow at first, so that some field research can be
performed. Smelters and beneficiation plants have been in existence in the
U.S. long enough that baseline data may be difficult to obtain, but steady
state concentrations in polluted natural waters can be investigated.
75
-------
TABLE 22. ESTIMATED RATES OF EMISSION OF CADMIUM DURING PRODUCTION
AND DISPOSAL OF CADMIUM PRODUCTS FOR 1968 (Ref. 70)
Item
Air contami-
nation
Water con-
tamination
Mining
and ore
concentra-
tion,
tons/yr
3000
Primary
cadmium
produc -
tion,
tons/yr
930
240
Electro-
plating,
pigment
and plastic
formulation,
tons/yr
300
Coal and
oil com-
bustion,
tons/yr
120
Cadmium -
plated
metals,
tons/yr
500
Pigment,
plastic s ,
and
miscellan-
eous*
tons/yr
90
Alloys
and bat -
teries,
tons/yr
40
Total
tons/yr
1680
3540
Soil contami-
nation
Accumulation
in service
Land disposal
(dumps, land
fills, slag
pits, mine
tailing s)
1420
2080
880
4390
300
310
360
500
490
220
2180
'Losses during use and disposal.
-------
Figure 23. Rates (tons/yr), routes, and reservoirs of cadmium
in the environment (Ref. 1Q).
11
-------
Although Cd appears to be resource limited clearly CdS solar cells
will see major use in pilot and demonstration plants in the near and inter-
mediate term (Ref. 14). Therefore, it is urgent that an appraisal of the
environmental effects of CdS production and cell manufacture be made
available.
We recommend that such a study- be performed soon for CdS and per-
haps GaAs. It may be premature, however, to place a high priority on
studies if InP, CdTe, and CuInSe2 manufacture. A study of Si manufacture
would be highly desirable even though Si is not identified as a pollutant. Si
halides used in vapor deposition could cause both atmospheric and waste
water releases if adequate control technology is not applied.
Avoidance of Coal Mining: We indicated in a previous subsection that
solar development might not replace coal fired utility boilers in the inter-
mediate term. Nevertheless, the potential exists if solar electric facilities
are able to achieve a market penetration which outstrips the opening of coal
mines and construction of fossil fuel fired electric plants.
Figure 24 depicts coal reserves in the western states. Most areas are
unsuitable for coal stripping, although very large coal stripping operations
MONTANA°H,.ena ?;»-)
Figure 24. Potential coal mining areas in the Western states (Ref. 77).
do exist or are being opened in Montana, North Dakota and Wyoming, parti-
cularly. This is a mixed blessing. Mining is the second most dangerous
occupation with an annual fatality rate of 71 per 100,000. Certainly, deep
78
-------
coal mining is more dangerous than surface mining. Avoidance of coal
mining will save lives and prevent crippling accidents. Industry figures
must exist to quantify this.
Other effects of coal mining and coal use now evident in the East which
will soon be common in the West:
• Disposal of overburden
• Erosion of slopes
• Landslides
• Mine acid drainage ("yellow boy")
• Coal haul road erosion ("red dog" roads)
• Gob pile leachate
• Gob pile burning
• Loss of topsoil
• Lost agricultural productivity
• Growth of low income shanty towns
• Siltation of surface waters
• Disturbance of ground waters
• Air pollution from coal cleaning plants
• Energy use in coal cleaning plant thermal dryers
• Coal fines lagoons
• Limestone stripping and quarrying for mine acid control
• Limestone stripping and quarrying for flue gas desulfurization
• Scrubber sludge disposal
• Scrubber energy consumption
• Precipitator energy consumption
• Fly ash disposal
• Bottom ash disposal
• Highway damage
• Diversion of resource from coal conversion
• Lost aesthetic values
Presently many utility plants in the Western states use natural gas.
Table 23 presents the fuel mix in the area. We know natural gas combustion
will not continue for long. These plants will either be converted to coal
(some to oil), or retired in favor of new nuclear, coal, solar, etc.
79
-------
TABLE 23. FUEL MIX AT ELECTRIC GENERATING PLANTS
IN THE WESTERN U.S. (REFS. 78, 79)
Region or Utility
Region, 1973:
West North Central
West South Central
Mountain
Pacific
Utility, 1975:
Central and South West
Gulf States Utilities
Ref.
Percent G
Coal Oil
eneration by Fuel
Gas Nuclear Hydro
and Other
Percent of fossil only:
78
78
78
78
60.8 2.2
3.2 6.0
60.2 8.4
5.8 47.0
37.0 — —
90.8 — —
31.4 — —
47.2 — —
All "fuels" considered
79
79
— —
— —
96 — —
96 — —
Houston Lighting and
Power 79
Oklahoma Gas and
Electric 79
Public Service,
Colorado 79
Southern Cal. Edison 79
Texas Utilities 79
62 —
15 40
22 —
96
99
33
15
77
20
Nationwide, 492.6 million tons/yr of new coal capacity are to be opened
by 1985 (Ref. 80). The mix is about 232x 10° tons deep and 284x 10° tons strip.
Different rounding and inclusion of auger mining and new mines opened in
1975 make the figures irreconcilable. The data are summarized in Table 24.
The names and locations of the mining companies, their mines, the destina-
tion of their products, and the year-by-year capacity expansion are all de-
tailed in Ref. 80.
It is obvious that even though a relatively small fraction of the areas in
Fig . 24 appear suitable for surface mining, stripping is the major expanding
segment of the industry, and steam coal usage is the major destination.
At some stage in solar development this expansion of coal mining may
be deterred or eliminated, or the coal will be released for use in conversion
plants^and chemicals industry. To the extent that this happens, lives will be
saved"" land will be saved, air and water pollution will be reduced, and a non-
regenerable resource will be released to more valuable uses.
One estimate is that 4,480 fatalities will have occurred in deep mining and
1,200 in surface mining of coal in the period 1970-2000 (Ref. 81).
80
-------
TABLE 24. NEW COAL MINE DEVELOPMENT IN THE WEST BY 1985 (REF. 80)
Arizona
Colorado
Kansas
Montana
New Mexico
No. Dakota
Texas
Utah
Wyoming
Deep Capacity
10^ tons/yr
—
10.2
—
—
—
—
—
29.5
3.3
Strip Capacity
106 tons/yr
8.0
13.6
0.8
55.5
17.6
30.1
19.7
—
120.9
Destination (10
Metallurg. Steam
— 8.0
4.5 9.3
— 0.8
— 55.5
— 17.6
— 30.1
— 19.7
— 29.5
— 110.2
tons)
Gasification
—
10.0
—
—
—
—
—
—
14.0
We believe it would be useful to assemble and collate an inventory of
fossil fuel fired plants in the regions hopefully to be the principal service
area for solar plants prior to the year 2000. This inventory should indicate
present fuel mix, suitability for conversion to coal, estimated retirement
date, heat rate, owner's plans for it and its replacement, etc.
Much of these data probably exist in EPA and ERDA files and in the
open literature. Such an inventory would be very useful in determining the
indirect environmental effects (beneficial and adverse) arising from new
energy sources.
Labor, Modular Construction, and Population Shifts: Project Independ-
ence (Ref. 23) figures for an accelerated photovoltaic production schedule are
given in Table 25.
TABLE 25. WORKERS IN PHOTOVOLTAIC PRODUCTION: PROJECT
INDEPENDENCE ACCELERATED SCHEDULE (REF. 23)
Year
1980
1985
1990
1995
2000
Capacity, MW ,,
Produced in
Year
5
100
1000
1000
1000
Workers
Inclusive of
Construction
1,932
38,700
430,000
1,720,000
3,440,000
81
-------
No separate estimates of construction workers is included. It would be
a useful number to have. A feature of photovoltaic utility systems is the
prospect of factory construction of modules. Thus, we do not anticipate large
population shifts to the Southwest due to device and panel fabrication. Popu-
lation shifts will occur wherever mining, smelting, materials production, and
panel fabrication occur. Population shifts to solar areas will be mostly due
to construction workers and operating personnel.
Additionally, there will be substantial demands for transportation
workers, construction workers, and municipal and business employees serv-
icing construction towns. The latter categories may be large. To analyze
the situation further is outside the scope of this report. Nevertheless, it is
a very important consideration which EPA, ERDA, HUD, FHA, DOT and
numerous other Federal agencies will want to study in a timely manner.
Transportation Demands: Increased transportation demands have been
alluded to previously. This area of concern is also outside the scope of this
report, but needs study. Siting of production facilities may need to be influ-
enced by transportation requirements as much as any other criteria.
Toxicology and Industrial Hygiene: An enormous labor force is en-
visioned according to Table 25. Large new commodity markets are to be
created. As a consequence of the known toxicity of some of the photovoltaic
semiconductor materials, extraordinary numbers of workers and families
•will be put at risk. The numbers involved probably dwarf the work force
potentially exposed to uranium and plutonium during the nuclear industry's
seminal years.
How prepared are we to deal with the industrial hygiene problems?
The famous Poison Control Center in Atlanta has nothing retrievable in its
files on gallium, indium and antimony (Ref.82). EPA is doing some in-house
work on indium, and has contracted with A.D. Little to prepare a report on
antimony (Ref.83). NIOSH has been active, also. In 1976 a new edition of
the NIOSH Manual of Analytical Methods will appear (Ref. 84). The following
list summarizes the analyses in the current edition and the expected 1976
edition.
NIOSH Manual of Analytical Methods (Ref. 85)
x = current edition
• = 1976 edition
Element (Analyte) Material (Matrix)
Blood Urine Air
Cadmium • • *x
Arsenic • «x «x
Gallium • •
Antimony • »x •
Indium • • •
Most methods will use anodic stripping voltametry.
82
-------
A reissue of the criteria document for arsenic is coming out (HEW 75-149),
with an atmospheric reference of 2 /lg/m , 15 min sample (Ref.85). A
cadmium criteria document is now in preparation. Fine particles analyses
and potable water analyses for some constituents of photovoltaic materials
have been published (Refs. 74, 75, 76).
A large literature exists on toxicology (and environmental toxicity) of
arsenic and cadmium. Gallium and indium are known to be toxic, in spite of
their positions below aluminum in the periodic table (Refs. 86, 87, 88). An-
timony was the subject of a very old review (Ref.89). To a surprising degree
the data are for laboratory animals rather than humans. Fleischer et al.
believe threshold values for cadmium are too low (Ref. 70).
Except for cadmium, and arsenic, little is known about the effects of
these materials on aquatic biota, nor is much known about low level chronic
inhalation toxicity on humans, domestic animals, and livestock.
If it appears that gallium arsenide is to receive much application,
strenuous efforts should be made to develop much better information on work
place safety and environmental toxicity. It is probably not too early to begin
such studies now.
d3
-------
SECTION V
CONCENTRATOR SYSTEMS
CONCENTRATOR TECHNOLOGY
The basic concept of concentrator technology is to collect solar radia-
tion over the land area required by the plant rating, and concentrate it onto
a manageably small area. The concentrated radiant energy is then used to
produce steam directly, or to heat a heat-transfer fluid, or to irradiate
photovoltaic cells. A heat transfer fluid may be used to convey energy to a
steam boiler; to supply process heat or space heat; or to store heat or con-
vey energy to a storage medium. Various combined concepts exist. We
will consider here only power conversion systems, and not process or space
heat systems. Two generic types of power conversion concentrator system
exist — central receiver and distributed.
In the central receiver, solar energy is transmitted optically from an
array of collectors to a central, tower top receiver ("power tower"). A
boiler may be located on the tower top; or on the ground. In the distributed
system, a heat transfer fluid must circulate from the tower top to the boiler,
or a reflector redelivers the radiant energy optically back to ground level
(Fig. 25). In distributed systems (Fig. 26), solar collectors are connected by
long piping runs (Ref. 90).
Collector concepts include fields ("farms") of sun-tracking flat mirrors
(" heliostats" ) (Fig. 25), tracking parabaloids or cylinders, fixed cylinders
(Fig. 26) or fixed Fresnel lenses, either circular or cylindrical. Cylindrical
receivers require the heat transfer fluid pipes to be arranged along and track
the linear focus of the reflectors, which sweeps across a cylindar arc in the
diurnal cycle. This is a relatively simple motion. Tracking paraboloids
are required to follow the sun in two degrees of freedom.
For useful operation of turbogenerators, temperatures exeeding 600 K
are desirable. Steam turbogenerators operating as low as 600 K exist; this
is comparable to the design temperature of pressurized water nuclear plants
(Ref. 90).
Tracking parabolic collectors can produce temperatures of 800 K. In
fact, concentration ratios of 1000X and more are achievable now using track-
ing parabolic dishes. Therefore, in principle at least, concentrator systems
may operate at the highest temperatures permitted by materials properties
and demanded by turbogenerator design. From the standpoint of land effi-
ciency as well as steam temperature, tracking paraboloids would be best.
However, there are experts who believe the system is too cumbersome, too
84
-------
100 MWe PLANT
BASELOAD
INTERMEDIATE
PEAKING
HYBRID
3 MODULES
(12 hr storage)
2 MODULES
(6 hr storage)
1 MODULE
(3 hr storage)
1 MODULE
(1/2 hr storage)
CHARACTERISTICS
TOWER HEIGHT
COLLECTOR AREA
AREA UTILIZATION
TOTAL LAND AREA
No. OF COLLECTORS
260 m
0.5 km2/MODULE
38.6%
1.3 km2/MODULE
15, 400/MODULE
SIZE OF COLLECTORS 32.4 m2
Figure 25. Solar thermal conversion: central receiver concept.
Schematic diagram of solar tower concept. The sun's
energy is reflected by heliostats to a receiver atop a
high tower and a coolant circulates through the receiver
bringing the reflected energy to the ground (Ref.40).
85
-------
CO
CTi
Figure 26. Fixed mirror solar farm (Ref. 36).
-------
delicate, and too difficult to control. They believe the tracking parabolic
collector is not a serious competitor (Ref. 36).
Serious objections have been raised also concerning the idea of install-
ing a steam boiler on top of a power tower. The authors of the Dow report,
for instance (Ref. 50) state:
"For those who have operated a high temperature boiler at
ground level, the thought of running a similar unit 500 ft
in the air is appalling. New maintenance systems will need
to be devised for such operations as replacing burned out
or plugged tubes. . . Just the mechanical suspension of such
a unit to permit human access without blocking radiation
appears to be a major challenge."
We might add that there could be substantial safety problems in per-
forming maintenance near the focus of a 100 MW field!
Two surviving central receiver concepts are thought to be especially
viable. In one, individually controlled tracking heliostats (flat mirrors)
focus onto a central reflector at the top of the power tower, which then re-
focuses the radiation onto a ground level converter. The other scheme re-
quires the hot heat transfer fluid to bring the energy to the ground. In both
cases, the steam boiler and turbogenerator would be ground based. Heat
transfer media may be steam or molten metals. Sodium has been proposed.
Table 26 presents some design dimensions for a power tower — collec-
tor form, which supplement those of Fig. 25.
The power rating implied in Table 26 may not be realistic, for it is not
likely that there will be 100% land coverage.
If we assumed 40% coverage and, 30% conversion efficiency, we get the
output values of Table 27.
The region receiving an average solar flux of 200 W/m includes San
Francisco, Salt Lake City, Denver, Little Rock and Atlanta.
Reflector materials may be glass, polished metal, or metallized plastic.
Much current research is devoted now to optimizing performance and reduc-
ing costs of reflectors. Indeed just as with photovoltaic generating station
concepts, cost reduction has assumed overriding importance. Fixed cylin-
drical or linear Fresnel reflectors (Fig. 26) may be composed of asphalt or
concrete onto which is impressed a metallized plastic or metal foil reflective
material. Clearly large quantities of reflector materials will be required to
satisfy any design requirements. These material requirements and other
data have been combined by one panel (Ref. 40) into an estimate that 100 mi
of solar power tower farm installed per year would correspond to one pro-
duction facility requiring three float glass lines and 1% of U.S. annual steel
production.
87
-------
TABLE 26. CENTRAL RECEIVER SOLAR POWER PLANT
REQUIREMENTS (REF. 36)
Item
Tower height (m)
Field diameter (km)
Equinox power (MWth)
x 30% — (MWe)
Number of mirrors:
144,000
64,000
36,000
16,000
7,000
4,000
1,800
800
Measurement
100
0.4
33
10
150
0.6
73
22
200
0.8
132
40
Mirror
300
1.2
293
88
450
1.8
660
200
600
2.4
1172
352
Size (meters)
Too Small
0.67
1.0
1.3
2.0
3.0
4.0
6.0
9.0
1.0
1.5
2.0
3.0
4.5
6.0
9.0
13.5
1.3
2.0
2.7
4.0
6.0
8.0
12.0
18.0
2.0
3.0
4.0
6.0
9-0
12.0
18.0
27.0
Too
3.0
4.5
6.0
9.0
13.5
18.0
27.0
40.0
Large
4.0
6.0
8.0
12.0
18.0
24.0
36.0
54.0
TABLE 27. CENTRAL RECEIVER SOLAR PLANT GENERATION
POTENTIAL (REF. 36)
Average Solar Flux
(W/m2)
260 (Highest)
200 (Average)
140 (Lowest)
Average Percent
of Possible
Sunshine
85
65
45
Average Power
Output
(MW/mi2)
91
70
49
88
-------
A further elaboration of the concentrator concept is exemplified by the
Sandia Laboratories combined photovoltaic-thermal electric system (Ref. Z3).
A linear collector mirror, either a parabolic or circular cylinder, concen-
trates radiation onto photovoltaic cells which are in thermal contact -with a
circulating fluid in a heat transfer loop (Figure 27). A silicon solar cell
loses output at a rate of about 0.5%/°C. Therefore, efficient heat transfer is
needed so that the combined converter actually achieves a net efficiency suf-
ficiently high to justify the added complexity.
Solar Collector
Heat
Loop
Solar Heat
Absorber
Thermal
Photovoltaic
Cells
Photovoltaic
Thermal
Figure 27.
Combined photovoltaic and thermal-
electric system (Ref. 23)
ENVIRONMENTAL IMPLICATIONS
Many of the environmental effects of concentrator systems have been
discussed explicitly or implied in the photovoltaic chapter. We will review
the situation here.
Siting and Grid Dispersal
Photovoltaic generating stations could be sized almost arbitrarily to
meet the requirements of specific load centers or of central generating facil-
ities. The principal constraint would be imposed by cost and design of peri-
pheral equipment. By contrast, concentrator conversion systems have eco-
nomies of scale which preclude small scale, grid dispersed application.
89
-------
Steam electric generating facilities are simply not suitable Tor modular
construction.
Even more restrictive is the water requirement. An exception is the
photovoltaic concentrator systems. All other concentrator systems require
a working fluid, usually steam and a cooling loop as well. Conventional wet,
mechanical draft cooling towers are frequently envisioned. Thermal elec-
tric efficiencies comparable to those of coal plants are expected: 30 to 32%.
Thus an ample supply of surface water is nearly essential.
Direct Effects
Surface Waters and Thermal Pollution: Heat rates comparable to con-
ventional fossil fuel plants imply thermal discharges of the same magnitude
for comparable plants. With these, we have blowdown, cooling tower chem-
icals, destruction of aquatic organisms at intake screens and in the cooling
loop, resultant BOD, salination, and the host of other tower effects. We do
not have coal pile drainage, fly ash and bottom ash disposal, scrubber sludge
leachate, coal fines lagoon drainage, etc.
Thus we avoid all of the feedstock related water pollution effects. On
balance, solar concentrator "farms" look very good indeed when ample sur-
face water is available.
Ground Water: The absence of toxic semiconductor materials is ad-
vantageous. It is possible that well water may be used for the cooling tower
makeup. Effects of groundwater extraction (such as subsidence) will need
site-specific investigation.
Land Use: Roughly speaking, land use requirements will be compar-
able to those of photovoltaic systems. We disregard differences of the order
of 50% as being comparable to variations occasioned by currently discussed
design alternatives. (See Figure 25 and Tables 26 and 27.)
Visual Effects: Power tower designs will have greater visual impact
than a flat plate photovoltaic facility. Masts of 200 to 400 m high are dis-
cussed seriously. These are comparable to utility fossil fired stacks.
Heliostat "farms" may create some glare which could be distracting when
viewed from higher elevations. Interference with vision of aircraft flight
personnel is conceivable and should be studied.
Terrain Effects: Earth moving and foundation preparation for the
power house and power tower will approximate the activities during con-
struction of a conventional fossil steam electric plant. The heliostat or
focusing reflector field will not require substantial earth moving.
Agriculture and Terrestrial Ecology: Because of the relative delicacy
of tracking mechanisms, and possible machinery hazards to livestock,
"multiple use" grazing is probably impractical. The shading effects dis-
cussed previously should be similar in the two systems. Shading, and hu-
midity increases from cooling tower drift and fog, may create proliferation
of previously uncommon flora.
90
-------
Air Pollution: Aside from cooling tower emissions no air pollutant
emissions can be identified in normal operation. Cooling tower drift will
carry chemicals.
If exotic heat transfer media such as molten sodium are employed,
accidental releases are possible. Ignition of a sodium spill would generate
quantities of caustic particulates and create a severe hazard to emergency
personnel.
Transmission Lines: This has been discussed previously.
Weather Modification: This has been previously discussed at length in
this report. Power tower installations will simulate greater surface rough-
ness than do photovoltaic or cylindrical distributed systems. Altered sur-
face wind velocities and cooling tower plumes will create greater potential
for microclimate change than previously discussed.
Noise: Unlike non-concentrating photovoltaic installations, concentra-
tor steam electric facilities will have rotating equipment. Noise levels
would be comparable to those experienced in conventional turbogenerator
applications.
Radiation: No ionizing radiation sources are anticipated.
Historical, Archaeological, and Paleontological Values: Because of
the delicacy of tracking equipment and personnel hazards, salvage archaeo-
logy probably cannot proceed after construction.
Accidents, Disasters and Sabotage: Seismic events, accidents, and
deliberate mischief could cause rupture of sodium lines, if this metal were
used for heat transfer. Ignition of the sodium could cause extensive per-
sonnel injury and equipment damage due to corrosive emissions and heat.
Combustion of storage materials is conceivable. Fire propagation could be
fierce if molded asphalt were used for fixed trough type collectors. Facility
security can be ensured, but with greater expenditure than is now the case
with more compact conventional facilities.
Resource Commitment and Depletion: No "exotic" materials are likely
to be needed in any substantial quantity. Thin film selective coatings for
reflectors are proposed. Gold and hafnium are candidates. A 1 GWe plant
might require one cubic meter of hafnium, or 1.5 x 10^ kg (Ref. 7) repre-
senting 46% of 1970 domestic production. However, hafnium production is
demand limited, not resource limited. World reserves at 1970 prices are
thought to be about 280 x 10° kg of which U.S. reserves are about 113 x 10b
kg (Ref. 62).
Energy storage requirements may be solved by pumped storage, since
surface water is likely to be available at a solar steam electric plant. A
hydrogen cycle system (discussed previously) is possible. But storage as
sensible or latent heat is seriously proposed. Heat transfer fluid may be
used. Another possibility is the use of phase change materials. Proposals
to use Na2SO4 -10 H2O and Na2S2O3 • 5 H2O (Refs. 91) are not attractive
91
-------
because the phase change temperature is much too low for turbogenerator
application.
Anhydrous Na2SC>4 has been proposed. A 1 GW plant would require
about 3 x 10^ kg of this substance for a day's storage capacity (Ref, 7), Pro-
duction of this quantity of Na2SC>4 would not strain our reserves. Discussion
of the vast variety of other storage materials is premature now. The inter-
ested reader should consult Ref. 92.
Indirect Effects: As for photovoltaic systems, much of the construc-
tion activity may occur in factories. Foil lined concentrator troughs cast in
concrete or asphalt are exceptions. No especially toxic materials are iden-
tified now. Until design features are better established, it seems premature
to discuss the environmental consequences of upstream manufacturing. Air
pollutant emission factors for Portland cement manufacturing, concrete
batching, steel, aluminum and glass manufacturing, etc., are all available
(Ref. 69) as are estimates of energy required to produce these rnaterials
(Ref. 68). When materials estimates become available for specific installa-
tions, or even conceptual designs, waste stream analysis and energy expendi-
tures may be calculated. (See pages 37 and 74 for discussions of possibly
beneficial delays or avoidance of coal mining.)
92
-------
SECTION VI
FLAT PLATE COLLECTORS
by P.O. McCormick
TECHNOLOGY ORIENTATION
Flat plate solar collectors have been in -widespread use since the turn
of the 20th century. Investigators have tried to take advantage of this abund-
ant, clean, "free" energy source for a variety of uses, including:
• Domestic Water Heating
• Space Heating
• Agricultural Application (Drying,
Curing, Water Desalination, etc.)
• Space Cooling and Dehumidification
Initially, the interest was principally in experimenting with a novel concept.
However, in the 1920s, the use of solar hot water heaters became popular
in areas where people wanted domestic hot water but lacked a convenient
heat source such as natural gas. This application was mostly in the South-
ern U.S. (and other areas of the •world such as Australia and Israel) where
sunshine was dependable year-round. Ironically, this use of solar energy
(at least in the U. S.) almost completely died out when alternate fuel sources
became readily available.
None of the other applications have ever had widespread use though
much experimentation has been done, especially in space heating and agri-
cultural applications. Solar air conditioning has recently been the object
of much research and a few demonstrations, but even the most optimistic
proponents see widespread solar air conditioning as being several years
away due to the high cost of heat driven air conditioning equipment (Ref. 93).
Demonstration projects are now getting underway to apply solar heating to
industrial process heat requirements and to agricultural product drying.
Current Technology
Current available flat plate collectors come in a variety of designs,
determined primarily by the intended application. The basic solar collector
that has been used since the turn of the century has a black painted metal
absorber with one or two transparent covers. Traditionally copper is used
for the absorber in liquid heating collectors while almost any metal is use-
able for air heating systems. Recently, there has been quite a substantial
93
-------
effort to make the absorbers of liquid heating collectors out of aluminum in
order to reduce cost. But, while the aluminum manufacturers were trying
to solve the corrosion inhibiting problem, steel absorbers have been making
inroads into the collector market. The transparent cover can be anything
that transmits light, but prevents outside air from blowing directly over the
absorber which would take the absorbed energy away. Many transparent
materials have been used, but glass tends to be the -material that is used
whenever a collector is designed to operate for any long period of time.
The next step up the development ladder for collectors has been the
introduction of selective coatings, a coating that absorbs well in the visible
spectrum but emits poorly in the infrared. The importance of selective
coatings becomes very pronounced as the operating temperature of the col-
lector gets above about 100 F (such as for solar air conditioning applica-
tions). A good selective coating can reduce the reradiation losses such that
collector efficiency is doubled compared to flat black absorbers at elevated
temperatures.
Convection suppression is also a means of increasing solar collector
efficiency at high temperatures. The so called "dead" air space between
the absorber and cover in a collector is by no means stagnant. Free con-
vection currents are set up by the temperature difference between the cover
and absorber. These free convection currents account for a large part of
the energy lost from the collector. Devices such as transparent and reflec-
tive honeycombs, (to physically interfere with free convection currents) as
well as evacuation of the airspace, are currently under development for con-
vection suppression.
For a review of residential installation technology and solar architec-
tural practice, see Steadman (Ref. 94) whose book contains illustrations of
numerous existing solar homes.
Effect of RfeD Efforts on Flat Plate Collector Use
Significant R&D activities are underway to increase flat plate collector
efficiencies, particularly at high temperatures. These concepts (such as
evacuated collectors, honeycomb convection suppressors, directionally
selective surface, and highly selective coatings) will influence the utilization
of solar collectors, particularly for solar cooling and process heat uses.
The cost/performance ratio of currently available solar collectors, along
with the $3000 price tag of a three-ton solar air conditioner, make resi-
dential solar cooling economically unattractive. With better collectors at
reasonable prices, solar cooling of homes could be economically feasible.
The lack of air conditioning capability is a substantial barrier to nationwide
utilization of solar energy in homes because many people who might install
a combined heating/cooling system, would not consider a heat-only system.
In any event, the major effect of R&D will be on the amount of solar
collectors used. The projected utilization rate if 0.37, 5.6 and 29.7 x 108
mz in 1985, 2000 and 2020, respectively, derived from Refs.4, 93, 95 and
96.
94
-------
Solar Collector Manufacturer^
There are many companies currently manufacturing solar heating
components. A comprehensive list of manufacturers is available in Ref. 97
(which is the result of an industry-wide survey by ERDA) and Ref. 98. The
following is an excerpt from these tabulations:
Corning Glass Works
E&K Service Co.
Energex Corp.
Energy Conservation Systems
Energy Systems, Inc.
Fred Rice Productions
Garden Way Laboratories
Grumman Aerospace Corp.
Halmac Co.
He lio - Dynamic s, Inc.
Intertechnology Corp.
Itek
Kalwall Corp.
Northrup, Inc.
Physical Industries Corp.
Powell Brothers, Inc.
PPG
Raypak, Inc.
Revere Copper & Brass, Inc.
Reynolds Metal Co.
Rodgers & MacDonald, Inc.
Shelley Radiant Ceiling Co.
SOL-R-TECH
Solarsystems, Inc.
Stolle Corp.
Sunearth, Inc.
Sunworks, Inc.
Tranter, Inc.
U.S. Solar Corp.
Unitspan Architectural Systems, Inc,
Materials Used in Flat Plate Solar Collector Systems
The assessment of -material uses for solar hardware must be largely
subjective at this time because three major questions are unanswered:
• Can the corrosion problems associated with using liquid
in a steel or aluminum absorber be solved to the satis-
faction of the users?
• Can integral collector/roof structures be made more
economically on site than collector modules made in
a factory?
• Will combined thermal/photovoltaic collectors be
employed in residential uses (cf. Fig. 6, para. 4.1.5)?
Liquid, rather than air, heat collectors are generally preferred be-
cause of pumping power difference, simplicity of heating system (ducting,
etc.) and the much easier application of liquid systems to heat driven air
conditioning systems. However, corrosion of liquid collectors is a very
real problem. Most current users of solar collectors are utilizing copper
absorbers (particularly those that are funded by the U. S. Government); this
makes collector panels very expensive. Either steel or aluminum absor-
bers would be less expensive but most people are reluctant to buy collectors
that may corrode in five years (as has happened many times). The solution
to this problem will give rise to the use of common materials such as steel
for the absorber.
The use of an integral solar collector/roof structure could, conceiv-
ably, reduce the collector materials to only that used in the absorber and
95
-------
cover. Amazingly enough, however, the only known research in this area
(Rcf. 99) is directed to producing prefabricated roofing panel/solar collec-
tors, not on-site construction of solar collectors as part of the roof structure
by building trades workers.
Based on these and other subjective auguments, the major materials
requirements (including all solar-related hardware) are 0.35 Ib AS., 3.0 Ibs
steel, 0.5 Ibs Cu, and 2.0 Ib glass per square foot of collector.
Storage Systems
The type and size of storage systems for use with flat plate solar col-
lectors is dictated by the particular use of the system. However, the follow-
ing general characteristics are typical of all uses:
• Easy addition and removal of energy,
• Small energy losses, and
• Capable of storing energy at or above
the temperature of use.
Historically, the preferred storage mediums have been water (with liquid
heating collectors) and rocks (with air heating collectors). The technical
merits of phase changes materials (PCM) and chemical change materials
(CCM) (Ref. 91, 92, 100) offer significant technical advantages over water
and rocks; their technical advantages never seem to outweigh their extra
cost, however. Undoubtedly, there will be some applications where flat
plate collectors will be used with either PCM or CCM storage, but the uses
appear to be so few that they are not addressed in the environmental assess-
ment. EPA may wish to watch the market penetration of PCM and CCM ma-
terials, but it would be premature to embark on a detailed environmental
assessment here.
The storage mediums that are expected to be of greatest use with flat
plate collectors between now and the year 2000 are:
• Treated Water; Water treated with 200-500 ppm of
corrosion inhibitor is the most versatile medium.
It will be widely used for applications in (1) domestic
hot water heating, (2) residential and commercial
space heating, (3) residential and commercial cooling
(both as hot and cold storage), and (4) industrial pro-
cess heat. For the process heat applications that re-
quire storage up to 120 C or so, a brine solution may
very well be the most cost effective storage medium,
even considering the corrosion problems. In many
climates, an antifreeze solution may be desirable.
• Rocks: The value of an air collector/rock storage
cannot be overemphasized when long life and de-
pendable service are of prime importance. How-
ever, the use of rocks normally makes a bigger
96
-------
impact on building space than other types of storage
and is, therefore, best used in new construction rather
than in retrofit. This is due to the larger size re-
quired for rock storage (compared to •water) and the
fact that air ducts are required rather than water pipes
to put energy in and take energy out of the storage.
ENVIRONMENTAL ASSESSMENT
Siting
The major contribution that flat plate collectors can make toward pro-
viding energy to replace conventional fuels is in residential applications.
This is not due as much to the energy requirements of space heating and
cooling as it is due to the temperature required for other applications.
Though there are many industrial processes that can make use of large quan-
tities of heat, most of them require temperatures too high to be supplied by
flat plate collectors (at reasonable efficiencies). Much of the process heat
that is within the range of flat plate collectors is supplied by degraded steam
at a very low cost per Btu. There are many legitimate uses for flat plate
collectors in industry (minerals, agriculture, food processing, etc.) but the
contribution of industrial applications to the use of flat plate collectors will
remain small. Since the large industrial facilities are not likely to use flat
plate collectors, there is not likely to be any large groupings of flat plate
collectors as might be expected of photovoltaic or solar thermal power pro-
duction.
There is the possibility that centralized collector farms may be used
to provide the heating and cooling for a community or, certainly, for high
density housing such as apartments and townhouses. Solar ponds, centrally
located in a community, have been proposed as a low cost solar heating sys-
tem. However, social problems and individual differences in comfort zones
will keep community solar equipment from becoming too popular. There is
also the technical problem of pumping energy (in the form of hot water, for
example) over long distances.
In summary, then it can be assumed within some limits that solar col-
lectors will be in use in direct proportion to the density of housing. Both
commerical and industrial uses will pose exceptions, but housing and com-
mercial buildings tend to grow up around industrial centers as a general
rule. Solar collector use inside large cities is an exercise in futility. Not
only is maintenance a difficult problem, but trying to ensure that solar col-
lectors will have a view of the sun for 20 years in the future will be difficult.
For large cities, it is much better to create power in a central solar elec-
trical plant and pipe it into the downtown area. Thus the heaviest application
for flat plate collectors will be in suburban or rural locations.
Direct Effects
It is difficult for an avid proponent of solar heating systems to admit,
even to himself, that there are potential environmental problems of solar
97
-------
heating, cooling and process heat applications.. Yet, the evidence is undeni-
able. People are willing to overlook shortcomings in a few, novel demon-
strations of solar technology, but when such uses become commonplace the
adverse impacts become evident. Some of the anticipated problems are:
• Contamination of groundwater by corrosion
inhibitors, algaecides and antifreezes
• Glare from collectors
• Interference from adjacent structures
• Danger of falling glass.
The lack of large scale use of solar collectors prevents quantitative assess
ment of these effects, but some general observations are made.
Contamination of Ground Water: The one serious drawback
to using water as a storage and heat transfer fluid is the
problem of corrosion. The problem is certainly solvable
but, based on current trends, can involve the use of chemical
inhibitors that are potentially environmental problems. The
best solution to the corrosion problem will likely be toxic
compounds such as the chromate family of inhibitors. The
inhibitor will get into the ground water from spillage during
installation and checkout, and from periodic maintenance
and corrosion.
Glare: Collectors that are tiltable such that they follow
seasonal variations in elevation of the sun will likely not
pose any glare problem to people on the ground. This is
due to the fact that most of the reflected energy will go
back into space rather than to the ground. Unfortunately,
it is difficult to design a variable tilt collector into the
roof of most house designs, so the majority of solar col-
lectors will be fixed in one tilt position. The optimum
tilt varies with latitude, weather conditions, and particu-
larly with use patterns, but nominal tilt angles will be 30
to 60 deg from the horizontal. This will definitely cause
glare on surrounding areas (particularly in summer) which
can be very offensive.
Some ways to minimize this glare are:
• Use a transparent cover that has a matte finish,
such as rolled glass,
• Install collectors, such that all the transparent
covers are not in the same plane. This will
break up the reflected sunlight.
Glare to vehicles in the air is not considered to be any
more of a problem than glare from swimming pools, etc.
The natural imperfections in flatness of the covers will
98
-------
break up the reflected light into a harmless glimmer for
aircraft more than a few hundred feet high. If an aircraft
is closer than that, the aircraft poses more threat to the
solar collectors than the solar collectors do to the aircraft.
A house near the end of an airport run-way where large jet
planes takeoff would be a good place to test the structural
integrity of solar collectors. It is anticipated that sound
from a jet airplane during takeoff would cause broken glass
and leaks in a collector.
Interference by Adjacent Structures: Most people think that
a solar collector needs a clear, unobstructed view of the
southern sky in order to function properly. While such a
view is desirable it is not necessarily required. It is pos-
sible for man to have his trees and solar heat, too. What
is required is judicious placement of trees and an occasional
pruning. Collector systems designed for heating and domes-
tic hot water are not significantly affected by early morning
and late afternoon shading. This is because very little energy
is normally collected before about 9 o'clock and after about
3 o'clock (solar time). Shading the house during the morning
and afternoon during the summer offers benefits in reduced
air conditioning demands which may outweigh the winter bene-
fits of no shading. Deciduous trees should minimize shading
problems, because they shed their leaves in winter, when
maximum heating loads occur.
So far, the discussion has been of the solar homeowner's
own trees or toolshed shading his own collector. Problems
are likely to arise when it is the neighbor's Lombardy popu-
lars that are doing the shading. Before, it was a question
of asthetics, but suddenly it becomes a problem that has en-
vironmental, legal, and social implications. The environ-
mental result may be a significant reduction in trees in
residential areas.
Danger of Falling Glass: Most solar collector installations,
particularly residential, will be on the roof (hopefully, most
new construction will make the collectors a part of the roof
structure). Until a better glazing 'material than glass is
found, a large amount is going to be put up on rooftop solar
collectors. Adequately stressed (such as with a big rock)
any glass will break, even tempered glass. Gutters might
catch most of the broken glass, but a lot of it is likely to
fall off the roof. This could pose real hazards to anyone
(but particularly small children) who happen to be in the
yard.
Natural forces, such as hail and high winds, can also cause
flying glass from collectors. Wind can pull large sheets
of glass from their frames (for example, the John Hancock
building) if the support is not properly designed and installed.
99
-------
At least a few solar heating systems will eventually be in-
stalled by almost every builder of homes in the United States.
There will be a learning period (and a fast buck, fly-by-night
period) during which quality of construction will be question-
able. Steps will have to be taken to prevent a serious problem
of falling glass by legislating quality standards and inspection
techniques.
On the positive side of solar energy, there are some environmental
problems that should be enumerated that will not be present with flat plate
solar collectors. Some of the more important ones are:
» Reduced fossil fuel consumption is the primary benefit
to society. Included is a reduction in all of the adverse
effects of coal mining and combustion and decreased
dependence on natural gas and imported oil. The pri-
mary benefit to the user must be reduced cost to heat
and/or cool his house.
« No air pollution as a direct result of solar collector use.
« Reduced heat imbalance. The solar collector causes less
increase in the net solar energy remaining on earth in the
form of heat than does an asphalt shingle roof, for instance.
• Few land use problems, since most houses have ample roof
space for collectors.
® Few adverse visual effects since one criterion for a builder
is to produce a home that will be acceptable to potential
buyers. A well built collector is not an eyesore; it is a
respectable, functional addition to a house.
Indirect Effects
Two major indirect effects of flat plate solar collector applications fall
under two major headings:
» The change in raw material uses due to the formation
of a new widespread product, and
9 The effects of forming a new industry which reduces
the market for an existing industry.
Raw Material Usage: As discussed previously, the actual materials to
be used in large scale solar collector production is still to be determined.
For purposes of this environmental assessment, however, an estimate was
made from the best available information and is presented in Table 28. The
data are shown as the amount of major materials used for the entire collec-
tor system. This includes plumbing, storage, heat exchangers and collector.
Many such tabulations appear to have omitted peripheral hardware. Even
though these use rates are significant, the impact on total production of these
materials will not be significant.
100
-------
TABLE 28. SOLAR HEATING AND COOLING
MATERIALS REQUIREMENTS
Cumulative Usage, 10 tons
through year
Material 1985 2000 2020
Aluminum
Steel
Copper
Glass
0.07
0.60
0.10
0.40
1.0
9-0
1.5
6.0
5.7
50.0
8.0
32.0
Industry Changes: The transportation costs for solar hardware, col-
lectors in particular, -would appear to stimulate regional manufacturing.
The relatively simple construction of solar collectors has already attracted
a multitude of small industries (sheet metal shops, for example) to develop
and market a solar collector. However, most of these industries are still
waiting for their first big collector order, and it is likely that a major por-
tion of the nation's solar collectors will be furnished by, through or for big
manufacturers. These big industries (such as PPG, Revere Copper, Dow
Corning, etc.) will simply expand their product lines to include solar col-
lectors, so the impact of solar hardware as a new industry will probably
not be large.
The impact of solar hardware on existing industries such as conven-
tional fuel heaters, will not be as great as might be expected. The nature
of solar energy is such that backup systems will almost always be required
for space heating and industrial process heating. Even solar air condition-
ing will have backup heating elements. However, due to the high initial cost
of some types of conventional systems (such as heat pumps, coal or oil furn-
aces) they probably will not be used as backup, to solar systems. Instead,
those of low initial cost systems, such as electrical resistance heaters, will
likely be used. It is highly unlikely, for example, that a conventional vapor
compression air conditioner will be used as a backup to a solar heat driven
absorption type air conditioner. Instead, a backup heat source for the ab-
sorption unit will be used.
Recommended Research
The direction that solar hardware development is to take during the
next decade is currently being shaped by government funding of R&D and
demonstrations. The environmental implications of these studies are
normally made a part of the study. However, some basic research needs
to be conducted on three topics.
• The environmental impact of various corrosion
inhibitors, algaecides, etc., should be investigated.
101
-------
Consideration should be directed toward establishing
the amounts of these chemicals that are expected to
be discharged intentionally or unintentionally.
* The extent of the problem of glare from collectors
should be assessed. Undoubtedly, glare can be
annoying; but will it cause accidents or other such
serious problems ?
* Tests should be performed on vulnerability to damage
and resulting human hazards from thrown objects,
sonic boom, seismic events, etc.
Summary
Table 29 contains a concise summary of some of the points discussed
in this chapter.
102
-------
TABLE 29. FLAT PLATE COLLECTOR SYSTEMS SUMMARY
Direct
Environmental Effects
Effects
• Corrosion Inhibitors
• Heat Transfer Fluids
, Millions of Ibs/year
1985
.2
.3
2000
3
4
2020
16
25
2. Glare
Ground to Air (No Appreciable Effect)
Ground to Ground (Possible Large Effect for Residential Use)
3. Reduced Consumption of Fossil Fuel
(Equivalent Gallons of Oil)
60x10 830x10 4,640x10
U)
Indirect Effects
Additional Cumulative Mineral Uses
Aluminum
Steel
Copper
Glass
(Millions of Tons)
07
6
1
4
1.0
9.0
1.5
6.0
5.7
50
8.0
32
Suggestions
1. Put out list of EPA recommended collector fluids and inhibitors.
2. Study reaction of glare from collectors by simulating collectors in urban community.
3. Study vulnerability to hazards.
-------
SECTION VII
REFERENCES
1. Energy Research and Development Administration. A National Plan
for Energy Research, Development, and Demonstration: Creating
Energy Choices for the Future Vol. 1: The Plan, ERDA-48, Washington,
B.C., June 1975.
2. Energy Research and Development Administration. A National Plan
for Energy Research, Development, and Demonstration: Creating
Energy Choices for the Future, Vol. 2: Program Implementation.
ERDA-48, Washington, D.C., June 1975.
3. Energy Research and Development Administration. National Program
for Solar Heating and Cooling (Residential and Commercial Applications.
ERDA-23A, Washington, D.C., October 1975.
4. Energy Research and Development Administration. Preliminary Defini-
tion Report — National Solar Energy Research, Development, and Dem-
onstration Program. ERDA-49, Washington, D.C., June 1975.
5. Office of Technology Assessment. An Analysis of the ERDA Plan and
Program. U.S. Congress, Washington, D.C., October 1975.
6. Prince, M. B. Photovoltaic Branch, Division of Solar Energy, ERDA,
Private Communication, February 1976.,
7. Dickson, E.M. Solar Energy. Control of Environmental Impacts from
Advanced Energy Sources. EPA-600/2-74-002, Stanford Research
Institute and USEPA, Washington, D.C., March 1974.
8. Griffith, L. A Position Paper on Solar Energy. Special Studies Staff,
IERL, USEPA, Research Triangle Park, North Carolina, August 1975.
9. U.S. Atomic Energy Commission. Proposed Final Environmental
Statement, Liquid Metal Fast Breeder Reactor Program, Vol. Ill, Alter-
native Technology Options. WASH-1535, Washington, B.C., December
1974.
10. Sholl, M. The MITRE Corporation, Bedford, Massachusetts, Private
Communication, February 1976.
II. Waliach, M. Lawrence Berkeley Laboratory, Private Communication,
February 1976.
12. Benson, J. W, Energy Research and Development Administration,
Private Communication, February 1976.
13. Gilmore, D., Environmental Protection Agency, Las Vegas, Nevada,
Private Communication, February 1976.
104
-------
14. Zeren, R.W. Electric Power Research Institute, Private Communica-
tion, January 1976.
15. Anon. Environmental Effects of Solar Energy to be Probed. Chemical
Engineering, Vol. 82, No. 27, 1975, p. 27.
16. Cherry, W.R. Agricultural and Process Heat Branch, Division of Solar
Energy, ERDA, Private Communications, January and February 1976.
17. Wolf, M. Photovoltaic Power. Astronautics and Aeronautics, Vol. 13,
No. 11, November 1975, p. 28.
18. Hickok, F. Handbook of Solar and Wind Energy. Cahners Publishing
Co., Inc., Boston, Massachusetts, 1975.
19- Killian, H. J., G. L. Bugger, and J. Grey, eds: Solar Energy on Earth
— An AIAA Assessment. American Institute of Aeronautics and Astro-
nautics, New York, April 1975.
20. Greene, G.U. Cadmium Compounds. Kirk-Othmer Encyclopedia of
Chemical Technology, 2nd Ed. Vol.3, Interscience, New York, 1964,
p. 899.
21. Doty, J. P., Eagle-Picher Company, Miami Oklahoma, Private Commun-
ication, February 1976.
22. Williams, J.R. Solar Energy Technology and Applications, Ann Arbor
Science Publishers, Inc., Ann Arbor, Michigan, 1975.
23. National Science Foundation and Interagency Task Force for Solar
Energy. Solar Energy Volume. Project Independence Blueprint Final
Task Force Report, Federal Energy Administration, November 1974.
24. Science and Public Policy Program. Energy Alternatives: A Compara-
tive Analysis, University of Oklahoma, Norman, and USGPO, Washington,
D.C., May 1975.
25. Szego, G.C. The U. S. Energy Problem, Vol. 2, Appendices, Part B,
NSF RANN 71-1-3, NTIS No. PB-207 519, Inter Technology Corp.,
Warrenton, Virginia, November 1971.
26. Szego, G.C. The U.S. Energy Problem, Vol. I: Summary. NSF-RANN
71-1-1, NTIS No. PB 207-517, InterTechnology Corporation, Warrenton,
Va., November 1971.
27. Sears, D.R. An Environmental Assessment of a 500 MW Hydrogen Air
Fuel Cell Peak Shaving Power Plant. LMSC-HREC TR D496563,
Lockheed Missiles & Space Company, Huntsville, Alabama, 1975.
28. Parrish, W. R., R. O. Voth, J. G. Hust, T.M. Flynn, C. F. Sindt, and N. A.
Olien. Selected Topics in Hydrogen Fuel. NBS Special Publication No.
419, National Bureau of Standards, Boulder, Colorado, May 1975.
29. Hausz, W., G. Leeth, D. Luechk, and C. Meyer. Hydrogen Systems for
Electric Energy, TEMPO Report No. 72TMP-15, General Electric Com-
pany, Santa Barbara, California, April 1972.
30. Post, R.F., and S.F. Post. Flywheels. Scientific American, Vol.229,
No. 6, December 1973, p. 17.
105
-------
31. Lawson. L.J., AiResearch Manufacturing Company of California,
Division of Garrett Corporation, Torrence, California, Private Com-
munication, Februrary 1976.
32. U.S. Atomic Energy Commission. Government-Wide Report to OMB
on Energy Storage R&D Program Strategies and Implementation Plans.
Division of Applied Technology, Washington, B.C., June 1974.
33. Fernandes, R.A. Hydrogen Cycle Peak-Shaving for Electric Utilities.
Paper presented at the 9th Intersociety Energy, Conversion Engineer-
ing Conference, San Francisco, California, 1974
34. Williams, J.R. Geosynchronous Satellite Solar Power. Astronautics
& Aeronautics, Vol. 13, No. 11, November 1975, p. 46.
35. Grosskreutz, J.C. Criteria and Procedures for Siting Central Solar
(Thermal)/Electric Generating Stations. Preprint of paper presented
at U.S. Sect. Meeting, International Solar Energy Society, Ft. Collins,
Colo., August 1974 (Black & Veatch, Consulting Engineers, Kansas City,
Missouri). See also: Site Selection Guide for Solar Thermal Electric
Generating Plants, Special Report prepared for NASA-Lewis Research
Center (Contract No. NAS3-18014) by Black & Veatch Consulting Engi-
neers, Kansas City, Missouri) June 1974; Selecting Preferred Sites for
'.. Solar Power Station Using Solar/Climatic Data, Special Report pre-
pared for NASA-Lewis Research Center (Contract No. NAS3-18014) by
Honeywell Systems & Research Center, 2600 Ridgeway Parkway,
Minneapolis, Minnesota, October 1973; and On-Site Survey of Candidate
Solar/Electric Power Plant Sites, Special Report prepared for NASA-
Lewis Research Center (Contract No. NAS3-18014) by Honeywell, Inc./
Black & Veatch, June 1974. (Available through Honeywell Systems fc
Research Center, Minneapolis, Minnesota.
36. Auburn University Engineering Systems Design Summer Faculty
Fellows. MEGASTAR: The Meaning of Energy Growth: An Assess-
ment of Systems, Technologies, and Requirements. CR- 120338,
National Aeronautics and Space Administration, September 1974.
37. See for example: Federal Register, 40 FR 55334, 55343, November 28,
1975, cf. esp. Section 130.17, p. 55341.
38. See, for example: (a) Jones, K. Hydrology of Limestone Karst in
Greenbrier County, West Virginia. Bulletin 36, West Virginia Geologi-
cal and Economic Survey, Charleston, West Virginia, 1973; and (b)
Bailey, L., and A.M. Malatino. Contamination of Ground Water in a
Limestone Aquifier in the Stevenson Area. Alabama Circular 76,
Geological Survey of Alabama, University, Alabama, 1971.
39. Thomas, H.E., Water. U.S. Dept. of Agriculture Yearbook. Govern-
ment Printing Office, Washington, D.C., 1955, p. 66.
40. See for example: Kimball, T. Is There an Inherent Conflict Between
Good Environmental Management and Energy Production? in Magnitude
and Deployment of Energy Resources, ERDA, Oregon State Board of
Higher Education, and Oregon State University, Corvallis, Oregon,
September 1975, p. 209.
106
-------
41. See for example: (a) Assessment of Environmental Impact of Alternate
Sources for Electrical Energy. Oak Ridge National Laboratory, Oak
Ridge, Tennessee, ORNL-5024, 1974; and (b) Robinson, N. Solar
Machine. Procedures, World Symposium on Applied Solar Energy,
Phoenix, Arizona, 1-5 November 1955.
42. Lodge, J. P., Jr., Chairman, Colorado Air Pollution Control Commis-
sion, and J. Pate, National Center for Atmospheric Research, Private
Communications, April 1973.
43. Reese, K.M. Strong Electric Fields Spur Growth of Chicks. Chemical
and Engineering News, 26 January 1976, p. 40.
44. Anon. Power-Line Radiation Affects Earth's Magneto sphere. Physics
Today, December 1975, p. 17.
45. Cermak, J.E. Air Motion In and Near Cities —Determination by Lab-
oratory Simulation: CEP 70-71 JEC27, Fluid Mechanics Program,
College of Engineering, Colorado State University, Ft. Collins, 1971.
46. Cormak, J.E., Director Fluid Mechanics Program, Department of
Civil Engineering, Colorado State University, Ft. Collins, Private
Communication, February 1976.
47. Auer, A. H., Natural Resources Research Institute, Laramie, Wyoming,
Private Communication, February 1976.
48. Holden, C. Hail Suppression Up in the Air, Science, Vol. 191, No. 4230,
March 1976, p. 932.
49. Atlas, D., National Center for Atmospheric Research, Boulder,
Colorado, Private Communication, May 1973.
50. Decker, G. L., R.W. Barnes, R. E. Sampson, and V. L. Prentice. Eval-
uation of New Energy Sources for Process Heat. Dow Chemical Com-
pany, Midland, Michigan, and Environmental Research Institute of
Michigan, September 1975.
51. Strickland, G., and J. J. Reilly. Operating Manual for the PSE&G
Hydrogen Reservoir Containing Iron Titanium Hydride. BNL 50421,
Brookhaven National Laboratory, February 1974.
52. Lueckel, W. J., and P. J. Farris. The FCG-1 Fuel Cell Power Plant
for Utility Use. Presented at the Summer Meeting of the IEEE Power
Engineering Society, 14-19 July 1974.
53. Manikowski, A. Lockheed Missiles & Space Company; Sunnyvale,
California, August 1975.
54. Gregory, D. P. A Hydrogen-Energy System. American Gas Associa-
tion, August 1972.
55. Dickson, E.M. Hydrogen as an Energy Carrier. Control of Environ-
mental Impacts from Advanced Energy Sources. EPA-600/2-74-002,
Stanford Research Institute and USEPA, Washington, D. C., March 1974.
56. Press Release No. 76-50, Marshall Space Flight Center, Huntsville,
Alabama, 5 March 1975.
107
-------
57 Su-vens Briscoe, Space Sciences Laboratory, Marshall Space Flight
Center, ITuntsville. Alabama, Private Communication, March 197b.
58. Sulyma, P.R., Lockheed Missiles & Space Company, Huntsville,
Alabama, Private Communication, February 1976.
59. Mumford, W.H. Heat Stress Due to R.F. Radiation. Procedures
IEEE, Vol. 57, 1969.
60. Michaelson, S.M., and C.H. Dodge. Soviet Views on the Biological
Effects of Microwaves -An Analysis. Health Physics, Vol. 21, July
1971, p. 108.
61. Glaser, P. Power from the Sun Via Satellite. Testimony before the
House Subcommittees on Space Science and Applications, and Science
and Astronautics, 24 May 1973.
62. U.S. Bureau of Mines. Mineral Facts and Problems. Bulletin 650,
1970 Edition U.S. Bureau of Mines, Washington, B.C., 1970.
63, Wade, N. Raw Material: U.S. Grows More Vulnerable to Third World
Cartels. Science, Vol. 183, 18 January 1974, p. 185.
64. Brobst, D.A., and W. P. Pratt (eds). United States Mineral Resources,
USGS Professional Paper 820, Government Printing Office, Washington,
D.C., 1973.
65. de la Breteque, P. Gallium. Kirk-Othmer Encyclopedia of Chemical
Technology, Vol. 10, 2nd Ed., Wiley-Interscience, New York, 1966,
p.. 3 11.
66. Donnay, J.D.H., G. Donnay, E.G. Cox, O. Kennard, and M.V. King.
Crystal Data, Determinative Tables, Second Edition. ACA Monograph
No, 5, American Crystallographic Association, New York, 1963.
67. Hammond, A. L. Lithium: Will Short Supply Constrain Energy Tech-
nologies ? Science, Vol. 191, 1976, p. 1037.
68. Anon. Report Estimates Energy Required to Produce Essential Pri-
mary Products. J. Air Pollution Control Association, Vol. 26, No. 2,
1976, p. 154.
69. Compilation of Air Pollutant Emission Factors, 2nd Edition. AP-42,
USEPA, Research Triangle Park, North Carolina, 1973.
70. Fleisscher, M. et al. Environmental Impact of Cadmium: A Review
by the Panel on Hazardous Trace Substances. Environmental Health
Perspectives, No. 7, May 1974, p. 253.
71. Inter society Committee. Methods of Air Sampling and Analysis.
American Public Health Association, Washington, D.C., 1972.
72, Bond, R.G., C.P. Straub, and R. Prober. Handbook of Environmental
Control, Vol. 1, Air Pollution, CRC Press, Cleveland, Ohio, 1972.
73. Natusch, D. F.S., J.R. Wallace, and C. A. Evans. Toxic Trace Ele-
ments; Preferential Concentration in Respirable Particles. Science,
Vol. 183, January 1974, p. 202.
108
-------
74. Hwang, J. Y. Trace Metals in Atmospheric Participates and Atomic
Absorption Spectroscopy. Analytical Chemistry, Vol.44, No. 14,
December 1972, p. 20A.
75. Lisk, D. J. Recent Developments in the Analysis of Toxic Elements.
Science, Vol. 184, June 1974, p. 1137.
76. Taras, M., A.E. Greenberg, R.D. Hoak and M.C. Rand, eds. Standard
Methods for the Examination of Water and Waste Water, 13th Edition,
American Public Health Association, Washington, D.C., 1971.
77. Sterba, J. P. Reclamation Plan for Strip-Mined Land Stirs Debate.
New York Times, 3 July 1974, p. 41.
78. Moody1 s Public Utility Manual, 1974, Moody1 s Investors Service, Inc.,
New York, 1974.
79- Bryant, A.H., III. Industrial Study and Appraisal: The Electric
Utilities. Reynolds Securities, February 1976.
80. Nielsen, G. F. Coal Mine Development Survey Shows 492.6 Million
Tons of New Capacity by 1985. Coal Age, Vol. 81, No. 2, February
1976.
81. Dials, G.E., and E.G. Moore. The Cost of Coal. Environment. Vol.
16, No. 7, September 1974, p. 18.
82. Hall, M., Poison Control Center, Atlanta, Georgia. Private Communi-
cation, February 1976.
83. Gruntfest, I. EPA Office of Toxic Substances, Washington, D.C.,
Private Communication, February 1976.
84. Crable, J.V., and D. G. Taylor. NIOSH Manual of Analytical Methods,
HEW Publication No. (NIOSH) 75-121, Cincinnati, Ohio, 1974.
85. Eller, P. National Institute of Occupational Safety and Health,
Cincinnati, Private Communication, February 1976.
86. Sax, N.I. Dangerous Properties of Industrial Materials, 4th Edition.
Van Nostrand, New York, 1975.
87. International Labour Office Encyclopaedia of Occupational Health and
Safety, McGraw-Hill and the International Labor Office, New York and
Brno, Czechoslovakia, 1974.
88. Christensen, H. E., T.T. Luginbyhl, and B.S. Carroll. The Toxic Sub-
stances List 1974 Edition. National Institute for Occupational Safety
and Health, Rockville, Maryland, 1974.
89. Fairhall, L.T., and F. Hyslop. The Toxicology of Antimony. Supple-
ment No. 195 to the Public Health Reports, U.S. Public Health Service,
Washington, D.C., 1947-
90. Gervais, R.L., and P-B. Bas. Solar Thermal Electric Power. Astro-
nautics and Aeronautics, Vol. 13, No. 11, November 1975, p. 38.
91. a. Telkes, M. Thermal Energy Storage. Record of the Tenth Inter-
society Energy Conversion Engineering Conference, Newark,
Delaware, August 1975.
109
-------
91. b. Telkes, M. Storage of Solar Heating/Cooling. ASHRAE Trans-
actions, Vol.80, Pt.II, 1974.
92. Hale, D.V., M. J. Hoover, and M. J. O'Neill. Phase Change Materials
Handbook. NASA CR-61363, Lockheed Missiles & Space Company,
Huntsville, Alabama, September 1971.
93. General Electric Corporation. Solar Heating and Cooling of Buildings
- Phase O, Final Report. Vol. 2, NSF-RA-N-74-02 IB, May 1974.
94. Steadman, P. Energy, Environment and Building, Cambridge Univer-
sity Press, Cambridge, England, 1975.
95. TRW Corporation. Solar Heating and Cooling of Buildings — Phase O,
Final Report. Vol. 2, NSF-RA-N-74-022B, May 1974.
96. Westinghouse Electric Corporation. Solar Heating and Cooling of
Buildings - Phase O, Final Report. Vol. 2, NSF-RA-N-74-023B, May
1974.
97. Energy Research and Development Administration. Catalog on Solar
Energy Heating and Cooling Products, ERDA-75, ERDA Technical
Information Center, Oak Ridge, Tennessee, October 1975.
98. Science Policy Research Division, Congressional Research Service,
Library of Congress. Survey of Solar Energy Products and Services
— May 1975. Committee on Science and Technology, 94th Congress,
June 1975.
99. Moore, S.W., J.D. Balcomb, and J.C. Hedstom. Design and Testing
of a Structurally Integrated Steel Solar Collector Unit Based on Ex-
panded Metal Plates. Los Alamos Scientific Laboratory report pre-
sented at the U.S. Section Meeting of the International Solar Energy
Society, Ft. Collins; Colorado, 20-23 August 1974.
100. Proceedings of the Workshop on Solar Energy Storage Subsystems
for the Heating and Cooling of Buildings, Charlottesville, Va., April
1975.
110
-------
APPENDIX A
ENERGY FORECAST BACKGROUND INFORMATION
The energy scenario analysis introduced in Section III contained very
little descriptive background information on these energy forecasts. The
assumptions and definitions of these five energy forecasts are presented in
detail in the following sections. The information presented is extracted
from Ref. 1.
SUPPLY AND DEMAND ASSUMPTIONS - SCENARIO 0 - NO NEW
INITIATIVES
Supply
• Oil and gas production draws on remaining recoverable re-
sources: (1) according to lower estimates by the U.S. Geo-
logical Survey (1975) and the National Academy of Sciences
(Ref. 1, p.IV-2), and (2) without tertiary or other new recovery.
• Coal and nuclear converter reactors continue to expand to
meet electricity demand, limited by ability to construct or
convert plants.
• Other energy sources (e.g., geothermal, hydroelectric, and
urban wastes) expand according to historic projections of
existing technologies which do not reflect recognition of a
serious energy problem.
Demand
• Current consumption patterns continue with no improvement
in residential, commercial, or industrial end-use and most
transportation efficiencies
• A 40% efficiency improvement for energy use in automobiles
is realized by 1980 because of a trend toward smaller autos.
SCENARIO I - IMPROVED EFFICIENCIES IN END-USE
;Jc
Supply
• Domestic oil and gas production is increased above the base
case (Scenario 0 ) by new enhanced recovery technologies.
'Other assumptions are essentially those of Scenario 0,
111
-------
• Solar heating and cooling are introduced.
• Geothermal heat is used for process and space heating.
• Waste materials are employed as fuels or are recycled
to save net energy in production.
o-
Demand
• Residential and commercial sector technologies are
improved with regard to:
— The structure itself in order to reduce heating
and cooling requirements
— Improved air conditioners, furnaces, and heat
pumps
— Appliances and consumer products.
« Industrial process efficiency improvements are
achieved in:
— Process heat and electric equipment
— Petrochemicals
— Primary metals.
• Efficiencies of electricity transmission and distri-
bution are increased.
0 Improved transportation efficiencies derived from
new technologies (in contrast to efficiencies from
smaller vehicles) are assumed for land and air
transportation.
0 Waste heat (e.g., from electric generation) is em-
ployed for other low-grade uses now requiring sepa-
rate energy input.
SCENARIO II - SYNTHETIC FUELS FROM COAL AND SHALE
Substantial new synthetic fuels production is introduced
from:
— Coal
- Oil Shale
— Biomass
Other assumptions are essentially those of Scenario 0.
The assumptions, unless otherwise stated, are those of the previous
scenarios to ensure that comparisons are being made only of the
impacts of stated energy options.
112
-------
• Enhanced oil and gas recovery levels of Scenario I are included
• Under-used solar, geothermal, and waste sources included in
Scenario 0 are not included here.
Demand
• No end-use efficiency improvements are assumed.
SCENARIO III - INTENSIVE ELECTRIFICATION
-.'c -jf
Supply
• Electric power is intensively generated by coal and nuclear
power as in prior scenarios
• New technology energy sources are introduced as available
to generate electricity
— Breeder reactors
— Solar electric (wind, thermal, photovoltaics and ocean
thermal)
— Fusion
— A minimal contribution is assumed from waste materials
(as in Scenario 0).
##
Demand
• Improved electric conversion efficiences are introduced
• Widespread use of electric autos begins
• Technologies to improve efficiency of electricity trans-
mission and distribution are implemented
SCENARIO IV - LIMIT ON NUCLEAR POWER
Supply
• Converter reactor energy levels are constrained to 200,000
megawatts electric
• Coal electric is at the levels in other scenarios to permit
coal to be employed for synthetics
'The assumptions, unless otherwise stated, are those of the previous
scenarios to ensure that comparisons are being made only of the
impacts of stated energy options.
'Supply assumptions are consistent with Scenario I and demand
assumptions with Scenario 0, unless otherwise stated.
113
-------
« Additional sources of electricity depend on
— Accelerated geothermal development (more
than a factor of two over Scenario III)
— Accelerated solar development (a factor of
two over Scenario III)
— Fusion as in Scenario III
e Solar and geothermal heating are used (as in
Scenarios I and III)
» Synthetic fuels are produced from coal, shale,
and biomass at the level of Scenario II.
Demand
• Industrial efficiency aspect of conservation scenario
(Scenario I) is included
« Electric transmission efficiencies are not included,
as electricity use grows too slowly to justify changes.
SCENARIO V - COMBINATION OF ALL NEW TECHNOLOGIES
Scenario V analyzes a case in which a combination of all major energy
packages, including nuclear, are simultaneously commercialized (i.e., im-
proved end-use, synthetic fuels, and electrification). The specific inputs for
this scenario are those previously summarized. It should be noted, however-
that the inputs are not simply additive; rather, potential energy supplies are
drawn on only as necessary to meet projected demand.
The senario results highlight the unbalanced impact of the total use of
technologies in meeting energy needs:
• A surplus of options for producing electricity is likely
to exist (e.g., neither coal nor nuclear options are hard
pressed to meet demand in Scenario V).
• Ability to meet liquid and gas requirements remains
marginal even if all current technological options are
vigorously pursued.
• Many technologies can complete to meet end-use needs
in some markets (e.g., utilities, industrial processes,
and space heating); few can compete in others (e.g.,
transportation and petrochemicals).
INPUTS FOR SCENARIOS
Quantitative energy supply and demand data which result from the pre-
ceding assumptions are summarized in the Table A-l.
114
-------
TABLE A-l. INPUTS FOR SCENARIOS (REE. 1)
Quantities
Item Year 1985 Year 2000
Scenario 0 — No New Initiative^
Electric Supply (GWe):
Coal 295
Nuclear —moderate growth
no LMFBR 185 7ZO
Hydroelectric — moderate
growth 86 92
Geothermal — expansion of
geysers 5 1Q
Oil and gas Remainder of demand*
Direct Fuels Production:
Oil (MBD) 10.1 5.3
Gas (TCF) 21.5 15.4
Coal As needed
Urban waste (Quads) 0.1 0.1
Consumption Technologies:
Automobile Efficiency (MPG) 17.5 20.0
Scenario I —Improved Efficiencies in End Use
Electric Supply Same limits as in Scenario 0'
Direct Fuel Supply:
Oil (amount added for
tertiary recovery) (MBD) 1.5 3.6
Gas (amount added for en-
hanced recovery (TCF) 5.0 7.4
Solar heating and cooling
(Quads) 0.25 3.5
Geothermal heat (Quads) 0.2 1.0
Waste material use (including
recycling) (Quads) 2.0 7.5
Waste Heat Use for Heat and
Power 0.4 3.0
"Amounts used less in some cases.
115
-------
TABLE A-l. (CONTINUED)
Quantities
Item Year 1985 Year 2000
Consumption (% improvement): Scenario I — (Continued)
Buildings:
-Shell 10 15
— Heating and cooling equipment 10 20
— Other appliances and consumer
products 10 25
Industry:
— Process heat and electrical
equipment 10 12
— Petrochemicals 5 25
— Primary metals 10 20
Electric power transmission
and distribution — 25
Transportation
— Land transport other than
autos 10 20
-Aircraft 15 15
— Autos (fleet average) 18.7 28
Scenario II — Synthetics from Coal and Shale
Electric Supply Same as Scenario 0
Direct Fuels Supply
Oil and gas Same as Scenario I
Synthetic crude and pipeline
quality gas from coal (or
equivalent barrels of oil) 0.7 7.0
Oil from shale (above ground
and in situ) 0.5 4.0
Biomass conversion (oil equivalent) 0.025 0.75
Solar and geothermal heat None
Urban wastes: Same as in Scenario I
116
-------
TABLE A-l. (CONTINUED)
Quantities
Item
Year 1985
Year 2000
Scenario II — (Continued)
Waste Heat Use and Electric Trans-
mission and Distribution
Consumption
Same as in Scenario I
*!*
Same as in Scenario 0
Scenario III — Intensive Electrification
Electric Supply (GW ):
Coal electric (max. in 1985)
Hydroelectric
Nuclear converter reactors
Breeder reactors
Solar elec. power
Fusion power
Geothermal elec. power
Oil and gas electric power
Direct Fuels:
C on s umption:
Same as Scenario II except
electric autos
295
86
225
0
1
0
10
not limited
92
720
80
50
1
40
Same as in Scenario I
10
Scenario IV — Limited Nuclear Power
Electric Supply:
Coal electric (max. level in 1985)
Hydroelectric power (same)
Nuclear converter reactors
Solar electric power
Fusion power
Geothermal
Oil and gas
295
86
185
5
0
20
not limited
92
200
100
1
100
Balance of demand
For all end-use efficiencies.
>'c ;'c
'Except waste materials and recycling added at base level.
117
-------
TABLE A-l. (CONCLUDED)
Quantities
Item Year !985 Year 2000
Scenario IV — (Continued)
Direct Fuels Supply:
Same as Scenario II (Synthetic
Fuels), plus:
— Solar Heating and cooling 0.25 3.5
- Geothermal heat 0.20 1.0
Consumption:
Same as Scenario 0, the
following efficiency improve-
ments from Scenario I:
— Process heat and electric
equipment 10 12
— Petrochemicals 5 25
— Primary metals 10 20
118
-------
APPENDIX B
CONVERSION FACTORS
1 acre
1 Btu
1 foot
1 inch
1 mile
1 pound
1 quad
= 4.047 x 10 square meters
_4
= 2.928 x 10 kilowatt -hours
= 3.048 x 10" l meters
= 2.54 x 10~ meters
= 1.609 x 103 meters
- 0.4536 kilograms
= 1015 Btu
1 sq. mi,
1 ton
1 Torr
= 180 million barrels of petrolem'
= 42 million tons of bituminous coal
o-
= 0.98 trillion cubic feet of natural gas '
= 293 billion kilowatt hours of electricity
= 2.59 x 10 square meters
= 9-072 102 kilograms
= 1 millimeter Hg
'£
These values vary with the quality of fuel actually
extracted and represent an average of recent production.
119
-------
TECHNICAL REPORT DATA
(Please read Instructions on //i.- reverse before
HI PORT NO.
EPA-600/7-77-0 86_
I TITLE ANDSUBTITLE
3. RECIPIENT'S ACCESSION-NO.
Preliminary Environmental Assessment of Solar
Energy Systems
b. REPORT DATE
Augus_t_1977 issuing date_
6. PERFORMING ORGANIZATION CODE
7 AUTHOFUS)
D.R. Sears and P.O. McCormick
8. PERFORMING ORGANIZATION REPORT NO.
LMSC-HREC TR D496748
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Lockheed Missiles & Space Company, Inc.
Huntsville Research Si Engineering Center
Huntsville, Alabama 35807
12. SPONSORING AGENCY NAME AND ADDRESS
Industrial Environmental Research Lab-Gin., OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
10. PROGRAM ELtMENT NO.
EHE624B
11. CONTRACT/GRANT NO.
Contract No. 68-02-1331,
Task Order 09
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/12
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Central station solar-electric plants and flat plate space heating installations
are environmentally superior to their respective conventional alternatives because
they produce little or no air and water pollution. Both kinds of installations will
require storage systems, also relatively clean environmentally.
Land area required for central station solar plants will be large, but it is not as
destructive or irreversible as with coal stripping. The ecological impact of solar
plants can be serious as a result of vegetation destruction. Visual effects can be
extensive, with no mitigating technology. Weather modifications may occur. Geo-
synchronous satellite generating stations could be environmentally catastrophic from
pollution caused by large numbers of space tugs.
Some photovoltaic materials, such as gallium and cadmium may be resource
limited. Indirect effects, resulting from the production of large quantities of photo-
voltaic materials, could be environmentally harmful.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Solar power generation
Solar energy concentrators
Solar heating
Environmental engineering
Pollution
Land use
environmental assess-
ment
IDA
10B
13B
13H
13 DISTRIBUTION STATEMENT
Release unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
132
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
120
U.S. GOVERNMENT PRINTING OFFICE: 1977- 75 7 -0 56 /6 51Z
------- |