EPA/AA/CTAB/89-07
Technical Report
An Overview of Photovoltaic and Battery Applications
by
J. D. Murrell
Karl H. Hellman
October 1989
NOTICE
Technical Reports do not necessarily represent final EPA
decisions or positions. They are intended to present technical
analysis of issues using data which are currently available. The
purpose in the release of such reports is to facilitate the
exchange of technical information and to inform the public of
technical developments which may form the basis for a final EPA
decision, position or regulatory action.
U. S. Environmental Protection Agency
Office of Air and Radiation
Office of Mobile Sources
Emission Control Technology Division
Control Technology and Applications Branch
2565 Plymouth Road
Ann Arbor, Michigan 48105
-------
Table of Contents
Page
Number
I. Introduction 1
II. Energy Uses and Fuel Sources 1
III. Solar Electric Power 3
IV. Energy Storage for Solar Photovoltaic Systems .... 8
V. Batteries for Electric Cars 9
VI. Other Vehicle Applications 11
VII. Conclusions 12
VIII. Bibliography 13
-------
I. Introduction
Solar energy is pollution-free, and politically and
economically stable. Its source does not, on whim, change its
availability or its price. Its reserves are essentially
inexhaustible and its users can collect it at no direct cost. Its
disadvantages are its diffuse nature and the fact that it is
sometimes unavailable in two ways: a predictable one (at night)
and an unpredictable one (due to cloud cover).
For these reasons, solar energy is attractive as an energy
source, and especially so from an environmental standpoint as a
potential replacement for the consumption of coal and petroleum.
Coal and petroleum generate significant quantities of particulate
and S02 emissions when burned in stationary uses, and petroleum
used in transportation vehicles generates large amounts of VOC
and CO emissions, even at today's levels of automotive emission
control.
II. Energy Uses and Fuel Sources
Two-thirds of all energy consumed in the U.S. goes to heat—
for stationary uses, or propulsion— for transportation; the
remaining third is consumed by power utilities to generate
electricity. The distribution of this energy use among major
user sectors is shown in Table 1, with liberal rounding for
clarity.
Table 1
Distribution of U.S. Fuel Energy by User (percent)
Consumed by Consumed by Utilities to
User User As Fuel Generate User Electricity
Industrial 25 10
Residential 10 15
Commercial 5 10
Transportation 25
TOTAL 65 35
-------
Accounting for the fuel mix consumed directly by the users,
and the mix of fuels used for utility power generation, and
apportioning that among the using sectors, the distribution of
energy fuels is as shown in Table 2, again with liberal rounding.
Table 2
Distribution of Source Fuel Energy
Among Using Sectors (percent)
User " Coal Oil Gas
Industrial 10 11 11 4
Residential 7374
Commercial 6253
Transportation 27
TOTAL 24 42 23 11
The intersection of Tables 1 and 2 for coal and oil is given
in Table 3. It shows that a new energy technology such as solar
would have the most impact on coal use if applied to electrical
power generation, and on oil use if applied to transportation.
Table 3
Coal and Oil Use, by
Purpose and User (percent)
Percent of Percent of
Coal Use Oil Use
Electricity:
Industrial 30 low
Residential 30 low
Commercial 25 low
Fuel:
Industrial 15 25
Residential low 5
Commercial low 5
Transportation low 65
TOTAL 100 100
- 2 -
-------
Although the industrial use of coal and oil as fuels is not
insignificant, most of it is for process heat, only half of which
is amenable to solar heating. The other half of process heat
energy needs could use solar thermal energy: about half of that
involves low temperature processes which could use simple flat
panel solar collectors; the balance goes into high temperature
processes which would require solar concentrators. Thus the
industrial sector's fuel use is not really very fertile ground
for simple (non-concentrating) solar thermal technology.
The use of oil for residential and commercial heating is not
very significant at the national level (10 percent of all oil
use). Energy for heating is significant to individual homes and
businesses, however, with space heating accounting for some 40
percent of their total energy consumption. Home and business
space heating energy comes nearly half from natural gas, about
one-third from oil, and most of the remainder from electricity.
Hence solar thermal space heating could provide consumer
benefits, but not much environmental benefit. It would reduce
the consumption of natural gas, making more of it available for
other purposes.
Ill. Solar Electric Power
Table 4 illustrates that large-scale electrical energy needs
could be met using modest fractions of land area, if solar energy
could be collected and converted to electricity at an overall
efficiency of 10 percent.
Table 4
Solar Power Area Requirements vs.
Electricity Needs. Large-Scale Areas
U.S.
Texas
California
Michigan
Rhode Island
Total Annual
Solar Energy,
Gigawatt-hrs*
15 billion
1.2 billion
790 million
170 million
3.7 million
Annual Elec-
trical Use,
Gigawatt-hrs
2.5 million
215,700
190,300
79,300
5,900
Solar Array Area:
% of
Total
0.17%
0.17%
0.24%
0.47%
1.58%
Acres
3.8 million
280,000
240,000
170,000
11,000
* horizontal flat plate
-------
Things Smaller Than Rhode Island
For smaller areas, electrical energy use is more
concentrated, of course. Table 5 is a worksheet which shows
typical annual energy requirements per unit floor space for five
types of buildings; note that the electrical energy consumption
density at this scale is in the 5 to 25 kwh/sq ft range.
Photovoltaic arrays sized to meet these buildings'
electricity needs would have areas of the same order of magnitude
as the floor space. Solar thermal collector areas sized to meet
heating needs are smaller.
In sunnier areas at lower latitudes, air conditioning
requirements will increase electrical demands, but energy
available from photovoltaic arrays will increase also; the
reverse is true for less-sunny locales at higher latitudes.
Table 5
Annual Energy Required Per Sq Ft Floor Space: Buildings
Type of Building Electricity. kWh/ft2 Heat. Btu/ft2
Residence 6.3 80,000
Warehouse 6.1 50,000
School 12.2 100,000
Wholesale/Retail 21.7 150,000
Hospital 23.9 200,000
Photovoltaic array size* needed to furnish electricity needs
(assuming 10 percent overall conversion and power conditioning
efficiency):
Residence 42% as large as floor space
Warehouse 41% as large as floor space
School 81% as large as floor space
Wholesale/Retail 145% as large as floor space
Hospital 159% as large as floor space
Solar thermal collector array size* needed to furnish heat needs
(assuming 80 percent collection/distribution efficiency):
Residence 20% as large as floor space
Warehouse 12% as large as floor space
School 24% as large as floor space
Wholesale/Retail 37% as large as floor space
Hospital 49% as large as floor space
2
* Based on a nominal U.S. solar insolation of 150 kWh/ft , or
512,000 Btu/ft2, on a horizontal flat surface.
- 4 -
-------
,. Table 6 illustrates this kind of regional variance for a
more or less typical residential photovoltaic system (one without
energy storage); it also illustrates how the purchase of night-
time energy from the power grid can be offset by energy generated
in excess of the house's daytime needs. It is worth noting that
a PV-powered house or business thus has a double benefit: it
does not contribute to daytime peak load demands on the grid, and
in fact, can furnish power into the grid at that time.
Table 6
Solar PV Residential Energy Balance, kWh/Year
(same Solar Array all sites, approx. 700 sq ft)
Fresno CA Madison WI Washington DC
Electricity requirements: 8,500 9,420 8,140
Electricity generated 12,710 9,420 9,440
Consumed by house 4,590 3,600 3,520
Sold to Utility, day 8,120 5,820 5,920
Bought from Utility, nite 3,910 5,820 4,620
Net bought/sold 4,210 sold even 1,300 sold
Things Smaller Than a House
Photovoltaic powering of a realistic car extends into the
range wherein power demand is too concentrated for the area that
can be used for solar panels, as shown in Table 7. Even using
solar power just for vehicle air conditioning is not practical.
Table 7
Vehicle Energy Requirements vs. Photovoltaic
Panel Capacity. Small Sedan (watt-hours/mile)
Electrical energy required for
vehicle propulsion: 300
Electrical energy required for
air conditioning— full capacity:
— slimmer- avrr-
65
— summer avg: 30
Electrical energy available* from
photovoltaic panel on roof:
* 100 watts/sq ft solar insolation, 10 percent conversion
efficiency, 20 sq ft available on roof.
- 5 -
-------
Although running cars solely by means of solar cells on the
cars shows no early potential for practicality, solar power for
cars can be quite feasible, by using battery-powered cars and
recharging them from stationary solar arrays at the home or
workplace. Table 8 shows that a solar array of reasonable size
can handle the energy requirements of a reasonable car operating
at a reasonable usage intensity. Note that the photovoltaic
array to run the car is about half the size of the one (Table 6)
to run the house.
Another way to power vehicles from a photovoltaic base
station would be to use the electric power to generate hydrogen
via the electrolysis of water, and use the hydrogen as a clean
fuel for combustion engines or fuel cells in the vehicles.
Table 8
Battery Electric Car and Solar Recharger
Vehicle Energy Requirement: 0.300 kWh/mile from battery;
0.460 kWh/mile into charger
5,040 kWh/year at 30 miles/day
Solar Array: at 150 kWh/ft2 insolation,
10 percent efficiency,
340 sq ft
Utility Compatibility and Cost
It would appear that the best niche for solar photovoltaic
power generation is electric utility use, and the best utility
niche is peaking power generation rather than base load power.
Peak power demand for utilities occurs in the summer in the
afternoon, just at the same time that the solar insolation, and
therefore solar array output, is also at a maximum. Some parts
of the country receive nearly 60% of their annual solar energy
during the five peak power demand months. Such an application,
without the cost burden of night-time energy storage, is a
natural match between solar photovoltaic power and utility power
needs.
Table 9 summarizes the magnitude of installed capacity of
U.S. utilities. The typical capacity of peaking power units is
that of new plants fueled with natural gas or petroleum; all of
the new gas- or oil-fired plants in 1987 were combustion turbine
or internal combustion units.
- 6 -
-------
Table 9
U.S. Electric Utility Plant Capacity. Summer megawatts. 1987
Total Capacity;
Avg. Plant Capacity:
Fuel
Coal
Gas
Nuclear
Hydro
Oil
Other
All
Plants
292,600
118,200
93,600
89,700
76,100
4,000
New
Plants
2,100
200
8,300
270
50
--
All
Plants
230
59
875
26
23
35
New
Plants
354
43
1,034
15
2
--
Capacity Range
New Plants
1.0 to 800
2.4 to 65
833 to 1,259
0.3 to 207
0.1 to 7
--
In order to compare the costs of a potential solar
photovoltaic powerplant to those of current powerplants, the
typical capital costs and operating costs for current peaking
power units were determined; these are given in table 10, with
costs for base load plants also shown.
Table 10
Utility Powerplant Costs
Fuel
Size
Capital cost
Operating and
Maintenance cost
Fuel
Total Operating
Peaking
Natural gas
50 mW
(gas turbine)
$300-$340/kW
1.355 $/kWh
2.52 $/kWh
3.88 $/kWh
Baseload
Coal
1000 mW
(two 500 mW boilers)
$1100-$1360/kW
0.429 $/kWh
1.52 $/kWh
1.95 $/kWh
- 7 -
-------
These are nominal values. Electricity costs vary
significantly across utilities. Some operating costs run much
higher, such as 27-53
-------
the battery charger from that system. The cost of such a system
would then depend on the cost of the charger, the batteries, and
the power conditioning needed to make the battery output usable.
Any excess power that couldn't be utilized by the then
current load or the battery could be sold back to the utility,
used for hot water heating, or other uses.
Using data from the EPRI Journal article and assuming that
the system needed to power an air conditioner overnight requires
about 30 kWh, then a storage system cost of roughly $3,000 would
be projected.
V. Batteries for Electric Cars
Two parameters of critical interest for batteries for
electric cars are: specific power and specific energy. Specific
power (watts/lb) can be related to the acceleration capability of
the electric car, and the range of an electric car is a strong
function of a battery's specific energy (watt-hours/lb).
One way to look at the capabilities of batteries is to use a
Ragone plot in which specific power is plotted against specific
energy. Figure 1 is based on the one in the JPL report entitled
"Should We Have A New Engine?", but we have modified it to show
where the gasoline-fueled conventional engine would be.
On the plot, higher and farther to the right is better. The
figure shows that there is a substantial difference between
batteries of different types and that none of the batteries shown
match the energy density of the conventional engine using
gasoline as the fuel. The log-log nature of the plot tends to
visually reduce the differences. For example, gasoline is
hundreds of times better than the lead-acid battery shown. It is
no wonder that battery cars accelerate slowly and have limited
range capability.
The batteries for electric cars can represent a substantial
portion of the car's cost, between 25 and 50 percent depending on
the battery type, according to Hamilton's report and the
Claremont report. A summary of some characteristics of batteries
considered potentially applicable is shown in Table 11. Most of
these values were excerpted from the Claremont report.
Batteries that are substantially better in one or more of
the performance indices in Table 11 would be considered
attractive candidates for electric car use. It should be noted
that some of the best current battery technologies, from the
standpoint of high watt-hour/lb capability, can release hazardous
and/or toxic materials into the environment. We are not sure
these problems are receiving sufficient attention by electric
vehicle designers and proponents.
- 9 -
-------
Figure 1
Ragone Plot: Battery Power Density vs. Energy Density
(Triangle Indicates Today's Gasoline-fueled Auto Engine)
1000
J3
I
i
100-
10
i-Clj
Ni-Cd
Pb-ACID
10 100
ENERGY DENSITY, W-hr/lb
1000
Source: JPL, "Should We Have a New Engine?
- 10 -
-------
Table 11
Battery Characteristics
Battery Type
Na-S
Li-Metal Sulfide
Ni-Metal Hydride
Zn-Cl
Zn-Br
Ni-Fe
Pb-Acid
Ni-Cd
Specific
Energy
fw-hr/lb^
41-75
36-64
33
27-32
25-30
20-25
20-23
11-26
Specific
Power
fw/lbl
45-80
45-64
84
36-41
36-50
27-47
36-91
kW-hr per
cu ft
3.1-9.9
3.1
2.1
2.4-2.8
2.4
2.5-3.0
120-130
90-130
60
95
75-80
125-400
70-95
300 plus
Safety/
Toxicity
Concern?
Yes
Yes
No
No
NO
NO
Yes
Yes
VI. Other Vehicle Applications
Solar Photovoltaics
Given that photovoltaic power tends to be low in watts
generated per square foot it is of interest to investigate how
photovoltaic power could be used for mobile applications.
One attractive application is for ventilation. When a car
is parked in the sun on a hot day, interior temperatures can
exceed ambient temperatures by a substantial amount. On a 100°F
day, interior temperatures can reach 120 °F or more. such hot
interior air sets the maximum design capacity point for
automotive air conditioning systems, since performance targets
are usually based on "pulldown," i.e., reducing the vehicle
interior temperature to a comfortable level in a short period of
time. If a way could be found to reduce interior heatup while
parked, then the air conditioner could potentially be downsized,
yielding cost and/or weight and/or fuel consumption benefits.
Solar photovoltaic ventilation units that can be retrofitted
to cars are available for sale today at a retail price of $30-35.
If they were to be integrated into the design of the air-handling
system of the car and produced in car-type production volumes the
cost would probably fall to somewhere in the $10 to $15 range.
Another possible application is for powering the vehicle air
conditioner itself. As shown earlier, solar photovoltaics cannot
provide enough power for current technology automotive air
conditioners; however, current belt-driven compressor air
conditioners were not designed to utilize solar photovoltaics, and
some air conditioner R&D could possibly yield more promising
results. The match between the problem (hot cars caused by sun
loading) and the possible solution (solar photovoltaics working
best under the same conditions) is too close a match to ignore.
- 11 -
-------
It is likely that solar photovoltaics could be used to
assist in battery charging. This could reduce the charging power
from the engine substantially, but not much has been reported in
that area.
Advanced Batteries
Advanced batteries have applicabilities in addition to the
obvious potential for electric car or hybrid car propulsion. An
advanced battery with greater power density could replace the
current starting, lighting, and ignition (SLI) battery and
achieve a weight savings. To the extent that the package could
be smaller, that would also be an advantage. Alternatively, the
same weight or volume could be used to provide more electrical
power or energy, and given the trends in increasing electrical
power demand for cars, this is the. route that we expect will be
taken.
VII. Conclusions
1. As concerns increase over greenhouse gas emissions due to the
consumption of fossil fuels to generate electricity, solar photo-
voltaic power will become more and more attractive.
2. Solar photovoltaic power has a natural match with peak load
electrical power demands caused by air conditioning usage.
3. The environmental benefits of solar photovoltaic power should
be assessed by comparing its negligible emissions to the emis-
sions of particulate, SOx, NOx, VOC, CO, and toxics from utility
peak load power plants, not base load plants.
4. Why aren't solar peaking power units in widespread use now?
Because no one has invested in high volume solar cell panel pro-
duction capacity to make economy-of-scale reductions in panel
costs needed for cost-competitiveness.
5. When considering advanced technology batteries for power or
transportation needs, the safety and environmental impact aspects
of the battery materials and designs may need closer scrutiny
than it appears is being given.
- 12 -
-------
VIII. Bibliography
The Claremont Graduate School (Hempel et al), "Curbing Air Pollu-
tion in Southern California- The Role of Electric Vehicles," Apr.
1989.
Electric Power Research Institute, "Technical Assessment Guide,"
Vol. 1, Dec. 1986.
Electric Power Research Institute Journal, "How Advanced Technol-
ogies Stack Up," July/Aug. 1987.
Electric Power Research Institute Journal, "Emerging Strategies
for Energy Storage," July/Aug. 1989.
Energy Information Administration, DOE, "Cost and Quality of
Fuels for Electric Utility Plants, 1987.
Energy Information Administration, DOE, "Electric Power Annual,
1987.
Energy Information Administration, DOE, "Monthly Energy Review,
1988 Annual Summary, Dec. 1988.
Intersociety Energy Conversion Engineering Conference, "Advanced
Energy Systems, Their Role in Our Future," Aug. 1984.
Hamilton, "Electric and Hybrid Vehicles, Technical Background Re-
port for the DOE Flexible and Alternative Fuels Study," draft re-
port, May 1988.
Hoff, "The Value of Photovoltaics: A Utility Perspective," IEEE
Photovoltaic Specialists Conference, May 1987.
Jet Propulsion Laboratory, "Should We Have a New Engine? - An Au-
tomobile Power Systems Evaluation," Aug. 1975.
Mazria, "The Passive Solar Energy Book," 1979.
Personal Communication, Energy Conversion Devices, Inc., Troy,
Michigan.
Photovoltaic Energy Technology Division, DOE, "Five Year Research
Plan: 1987-1991," DOE/CH110093-7, May 1987.
Photovoltaic Energy Technology Division, DOE, "Investing in Suc-
cess," Nov. 1988.
Society of Automotive Engineers, "Recent Advances in Electric Ve-
hicle Technology," SP-793, Aug. 1989.
Solar Energy Research Institute, "Toward a New Prosperity," 1984.
- 13 -
------- |