as today, many scientists and technicians were afraid to publicly announce their opposition to nuclear power for fear of recriminations by the AEC or the nuclear industry. Then, as today, solar power advocates were for the most part critical of nuclear power because of its disastrous environmental effects.

In March 1959, the University of Maryland's Bureau of Business and Economic Research published a prescient document, Solar and Atomic Energy—A Survey, which assessed the future effects of the development of nuclear reactors vis-à-vis the development of solar power. It concluded:

Modern industrialized society is now in possession of a concentrated atomic power which could destroy it. In the history of science, development of nuclear fission was a notable accomplishment. The thrill of its potentials and the challenge of its application have influenced the planning of the military and the allocation of funds in the Nation's budget. . . . Application of atomic energy to peaceful uses is under test, it is true; but the emphasis is minor, and the basic dangers of contamination of the human race through radioactive waste, medical overdose of isotope therapy, and accidents, particularly in transportation, are terrifying.

Fortunately, there is an alternative source of energy waiting to be developed, and an international community interested in its application to the welfare of mankind. This is solar energy. . . . Now, before our conventional fuels are in short supply and we and our environment are more contaminated by radioactive substances, is the time for solar research.

John H. Cover, an economist who at that time was serving as director of the Bureau of Business and Economic Research, recalls clearly the hostility with which the report was met. "I had the advantage of a number of scientific and technological consultants," he explained recently, "but was dismayed that only two would permit reference to their participation, although each read and all approved of the manuscript." *

*

(2.) "The Promise of Mass Production"

The promise of mass production

Despite the fact that cost considerations are not the only-perhaps not even the major-factor in deciding whether solar or fossil or nuclear fuels should provide future electrical power, there are many who are convinced that, with further research, solar power can nonetheless become as cheap as or cheaper than either fossil or nuclear power. Certainly a key determinant in bringing this into being will be the development of mass-production techniques.

William Cherry, an engineer who has worked with solar cells since the 1950s and is now a NASA official at the Goddard Space Flight Center in Greenbelt, Maryland, has advocated a combination of new techniques to mass-produce solar cells for use in power plants in the United States. He foresees an efficient, highly automated assembly line for the production of the silicon cells, in which one end of the machine is being fed a roll of silicon raw material while from the other emerges a finished solar cell "blanket."* Using sophisticated mass-production techniques, solar cells would cost about $0.50 per square foot, or $50 per kilowatt, he has said.

A machine very similar to that proposed by Cherry has been developed for mass-producing "ribbons" of high-quality sapphire for various technological applications. The developers of the sapphire-ribbon machine are engineer Harry E. Labell, Jr., the inventor, and Dr. A. I. "Ed" Mlavsky, director of the Corporate Technology Center of Tyco, Inc., in Waltham, Massachusetts.

The Tyco mass-production process is based on the principle of capillary action. At a blistering 2,620° C, molten sapphire is drawn through a capillary tube. When it reaches the top of the tube, a specially designed "die" made of sapphiremolybdenum causes the sapphire to conform to its shape-i.e., square, circular, rectangular, etc. Another solid piece of sapphire is attached to the molten die, and then pulled away, bringing with it the molten sapphire from the capillary tube in the shape of the die. Additional sapphire rises in the capillary tube, replacing the sapphire drawn through the die and continuing the process. A sapphire ribbon

*The two men who signed the final report were scientist/author Ralph Lapp and engineer Ellery Fosdick.

*The violet cell technology was sold by COMSAT in 1973 to the Centralab Division of Globe-Union. Inc., one of the two American silicon space solar cell companies. COMSAT and NASA have announced that violet cells will be used on future satellites.

is produced by drawing the molten sapphire through a specially designed gap, or openinng, in the die. The gap can be almost any shape, so that the sapphire ribbon may emerge cylindrical, as a rectangular strip, or in some other desired configuration.

In a 1972 interview, Dr. Mlavsky outlined preliminary research and development work aimed at producing silicon ribbons in the same way that Tyco now produces the sapphire ribbons. The temperature of molten silicon is lower than that of sapphire, which could eliminate some of the problems attached to sapphire mass production. (At 1,420° C, silicon's melting point is more than 1,000° F lower than that of sapphire.)

However, silicon poses other additional difficulties, the principal one being its reactive chemical nature, which causes the dissolution of many elements and the formation of new chemical compounds with others. This creates substantial problems in the choice of a suitable die material that could withstand the rigors of continuous mass production of silicon strips. However, Dr. Mlavsky said that preliminary efforts were quite encouraging, and predicted that the problem could be solved-particularly through the application of an advanced development program and several hundred thousand dollars in financing.

By 1973, prototype machines were producing crystalline silicon ribbons on a small scale, but-as feared-some reaction problems with the die materials were preventing the silicon ribbon from achieving a quality high enough for use in solar cells.

Nonetheless, the Tyco process has raised the hopes of solar power advocates that the automation problems can be solved so that the costs of finished silicon cells will drop dramatically.

Dr. Mlavsky has estimated that, on the basis of future raw silicon costs of $10 per pound, a small factory of only 12 machines could continuously produce 2,400 1-inch-wide ribbons of silicon pure enough for solar cells at a cost of $17.50 per pound of solar cell ribbon. This would bring the cost of silicon per kilowatt of solar electrical power produced to $90. Of course, this represents only about half the cost of the finished solar cell, which Mlavsky conservatively anticipates could be built for $180 per kilowatt utilizing these new techniques.

Dr. Mlavsky believes that the successful development of the Tyco mass production process would make possible the future use of silicon as an energy "storage" material-i.e., as a transportable fuel. Silicon ribbons-suitable for fabrication into solar electric cells--would be manufactured in a location having available sand and electric power (which could come from a solar plant). After production, the ribbons, or perhaps the completed silicon solar cells, would be shipped to their ultimate destinations for use in converting sunlight to electricity. In a sense, this would be similar to the present-day shipment of gasoline, oil, gas, or the transportation of electricity in power lines. Dr. Mlavsky believes that silicon shipment would be more advantageous economically and environmentally than conventional energy transport.1

at

A coalition of solar cell scientists from the solar cell industry (with one representative from the government-funded Jet Propulsion Laboratory, Pasadena, California) presented a technical paper on the economics of solar cells in 1972, arguing that the Tyco process could provide solar cell silicon "blanks" (the pure silicon slices on which the additional electric conductor layers are deposited) a cost of $250 per kilowatt, more than a hundred times cheaper than the costs of cutting silicon from furnace ingots-$30,000 per kilowatt. "The development of a continuous ribbon growth technique is the key necessary for more economcial silicon solar cells," they said. From this, they extrapolated that a production line for finished solar cells (as William Cherry suggested) could become a reality, in which all the tasks of making the cell's electric junctions, conductors, and coatings would be automated. They concluded that at 10 percent cell efficiency. the costs of electricity would be about $375 per kilowatt of plant capacity at a

1 "Take Africa, for example," Dr. Mlavsky said. "There is the Sahara Desert as well as an abundant supply of cheap hydroelectric power. The ribbon would be made from sand on site, using hydroelectric power to heat the furnaces (to make the ribbon). It's a completed loop system where silicon is used as an energy storage medium. Silicon is portable. One truckload of silicon (in completed solar cells from ribbons) would generate many, many thousands of kilowatts." Dr. Mlavsky added that this sort of energy storage concept should be compared to the production of hydrogen fuel from solar energy, referred to at length in a subesquent section. His argument is that silicon has more energy content capability and is cheaper to transport than hydrogen fuel.

2 The Tyco process would produce a continuous ribbon, which would be automatically cut into blanks for the cells a much simpler process than cutting silicon ingot slices, and one not requiring precision diamond saws.

*

sunny location. This would bring the silicon solar cell to commercial realization, since today's conventional power plants cost $300 to $600 per installed kilowatt. The additional costs of storing solar power for nighttime use would have to be taken into account, but this cost would probably not double the installed power costs. With mass production of silicon solar cells, solar power could become competitive with nuclear power plants.

Representatives of the Heliotek-Spectrolab Division of Textron, Inc., have attempted for several years to encourage federal and private funding of advanced technologies to improve the performances and lower the costs of silicon solar cells. Heliotek-Spectrolab engineer Eugene Ralph in 1970 and 1971 advocated a solar cell development program that would lower the cost of solar cells for Earth power from $100,000 to less than $15,000 per kilowatt.

Ralph's comprehensive plan would initially consist of redesigning the shape and composition of NASA-quality silicon cells. The first step would be producing the cells-not in squares and rectangles required for satellites, but round, in the shape of the silicon ingots from which they are cut. This would eliminate the waste of energy-intensive silicon, which is discarded as the square cells are cut from the round ingots. This one achievement, coupled with mass production, he said, would be sufficient to produce the projected cost cut from $100,000 to $15,000 per kilowatt of electricity.

Additional savings could be achieved by using mirrors or optical lenses to concentrate more photons of sunlight on the solar cells, making possible more power output from them.* Under Ralph's direction, the Heliotek-Spectrolab laboratory produced special concentrator solar cell arrays, called "egg carton" concentrators, because the solar cells are placed at the bottom of mirror-lined cavities that look like the inner walls of egg cartons. This concentrating device produced 21⁄2 times the electric power output of similar cells without concentrating mirrors.

Ralph believes that much higher concentration ratios will increase the power output, while lowering costs still further. One experimental approach he has developed would increase the power output of a cell equipped with mirrors 125 times above that of the naked solar cell.

"The commercial solar cell cost of $15,000 per kilowatt would then decrease to about $150 per kilowatt," he predicted, “which would not only far exceed the ... cost goals, but . . . would bring solar energy conversion into a competitive status with conventional fossil fuel power generation stations."

(3.) Conclusion: Large-Scale Solar Power Projects

Conclusion: Large-scale solar power projects

The prospects for developing solar power on a large scale are indeed intriguing, but the promises and aspirations of a few scientists may be fraught with errorunintentional, but potentially disastrous in terms of social planning. The primary flaw in most proposals for harnessing solar power is that the proponents have not accurately accounted for the amount of energy required to build and maintain solar power plants. In the preceding analysis of solar bioconversion projects, Dr. Howard Odum's comparison of the energy needed to build and maintain complex laboratories for synthetic fuel production, compared to the fuel delivered from such schemes, is a clear analysis of possible technical miscalculations.

What is needed from the proponents of other solar technologies is a clear accounting of the energy required to build and operate plants using the sea-thermal, photovoltaic, or large-scale solar heat plants. Research should be quickly initi ated to evaluate and assess the potential to society from each solar process, rather than relying on faith in corporate wisdom to to select a rational alternative power process for society.

The levels of government funding for solar energy have never been great, and the few million dollars that the government now allots for solar research is clearly inadequate to the task ahead. If our society is to reap any benefit from solar energy in the power production area, an industry-government coalition should begin work now to avert the real possibility of social collapse when the fossil fuel levels decline and the only energy source available is nuclear fission.

*Some of the problems with mirror concentration are discussed in this chapter in the section on a space solar power system.

56-516 O-75 Pt. 1C 52

Pipes and fittings-$100 plus an additional 10 cents times the number of square feet of solar collector area.

Motors and pumps (for circulating water or forced air)-$50 plus 20 cents times the number of square feet of solar collector area.

Heat exchangers (for transferring solar-heated water in the house)-$75 plus 15 cents times the total square feet of collector area.

The cost of the solar collector were estimated in two ranges: (1) the estimated costs of solar collectors manufactured today in factor lots suitable for installation-concluded to be $4 per square foot of area; and (2) the cost of a solar collector manufactured on a truly mass-production basis to fulfill the demands of a future upsurge in orders for solar-heated homes-estimated to be about $2 per square foot of solar collector.

The 8 U.S. locations considered were selected as representative of the various climate zones: Miami, Florida-tropical savannah climate; Albuquerque, New Mexico-tropical and subtropical steppe; Phoenix, Arizona-tropical and subtropical desert; Santa Maria, California-Mediterranean or dry summer subtropical; Charleston, South Carolina-humid subtropical; Seattle, Washingtonmarine West Coast; Omaha, Nebraska-humid continental, warm summer; Boston, Massachusetts-humid continental, cool summer.

Results of the study indicated that in 6 of the 8 cities surveyed a good solar house system would provide cheaper heat than electrical heat at prevailing power rates. In only 2 cities, Miami and Seattle, would economics favor electricity over solar energy for home heating. In Miami, it turns out that home heating is not required enough of the year to warrant the high capital costs of solar space heating systems; and in Seattle, very cheap electricity available from the government-subsidized hydroelectric dams of the Pacific Northwest make solar heating uneconomical in comparison.

On the other hand, the solar heating systems could not compete effectively with conventional home furnaces fueled with oil and gas at prevailing fuel costs, so that solar heating systems were found to be cheaper in only a few of the locations studied.

However, because the costs of oil and gas have risen appreciably since the study was made, even this economic indicator has changed, and solar home-heating systems might be economically competitive in more areas today.

The study was made to ascertain the economic feasibility of solar energy-not necessarily what the most effective solar heating system would be, but what the least expensive and most economically competitive would be. This aspect of the study affected the design objectives of the houses. For example, the heat-storage requirements covered only 1 to 2 days of winter demand, and the entire solar collector heating system was designed to supply about half of the house heating needs in most locations. In only a single very favorable location-Santa Maria-would the small solar heating system supply three fourths of the winter heat requirement, using a relatively small collector of 261 square feet. In Charleston, a 208-square-foot collector would supply 55 percent of the heating need: and in Omaha, a 521-square-foot collector would supply 47 percent of the heating needs.

b. Excerpts from Chapter 7, “Decentralized Uses of Solar Energy”

Today, many engineers and scientists working in the field of solar power for decentralized applications maintain that not only are new solar house designs sound from the standpoint of engineering, but from the standpoint of economics as well.

Two comprehensive studies comparing costs of solar-heated houses with costs of houses heated with conventional fuels have been conducted by Dr. George Löf in collaboration with economist Richard Tybout, professor of economics at Ohio State University. A preliminary study published in 1970 on house heating economics was expanded, and in 1973 the authors published the results. The sophisticated survey, based on the performance of solar heating systems in a number of U.S. locations showed that solar heat is economically practical now in most

The survey utilized more than 400,000 hourly recorded observations of solar radiation, temperature, wind regime, cloud cover, and humidity in each of 8 U.S. cities. The data were programmed into a computer and matched against performance capabilities of solar collectors of known design and their related heat-storage systems, along with the energy demand, insulation requirements, house size, and other important factors involved in designing solar-heated dwellings for each of the specific geographic areas.

The study made two major contributions: (1) optimization of the economic design of solar heating systems; i.e., the amount of space required for the collector and for storage, and the collector orientation that would be most economic in each location; and (2) establishment of realistic costs of solar heating vs. recorded costs of conventional fuels used in house heating today.

The study compared the costs of solar heating systems supplemented with electricity and with conventional oil and gas-fired furnaces, to the costs of nonsolar-supplemented electric, oil, and gas-heating systems. Costs of the solar heaters were based on amortizing the high initial costs at 6 percent interest over a 20 year period.1

The initial costs of the solar heating systems were estimated for 2 different sizes of home: (1) a large house in the "upper-middle-income" bracket, with a heat demand of 25,000 BTUs per degree-day of heat; and (2) a house in the "middle-income" range, with a heating demand of 15,000 BTUs per degree-day. Controls for the solar heating system were estimated to cost a total of $375, exclusive of a conventional house heater (used as an auxiliary with the solar heating system) and associated heat distribution pipes or ducts.

These costs are broken down as follows:

Water storage system-one-half cent per pound of water stored multiplied by the number of square feet of solar collector space used.

Controls (thermostat and associated equipment)-$150 for all systems regardless of size.

The authors found that only a single glass plate would be needed to cover the solar absorber plate in the Phoenix and Miami locations, but 2 plates would be required in the other cities.

They said that "it is probable that solar heating costs will decrease somewhat as improvements are made," and that "competitive solar heat will become increasingly possible as these trends [the spiraling costs of conventional fossil fuels] continue."

They further concluded that:

Conditions conducive to economical solar heating are moderate-to-severe heating requirements, abundant sunshine, and reasonably uniform heat demand during the period when heat is needed. The higher the cost of conventional energy for heating, the more competitive a solar-conventional combination becomes.

The value of this economic survey is that it emphasizes the principle of lifecycle costing-i.e., the cost and maintenance of materials that go into the house (including the heating system) should be amortized over the lifetime of the house. Today, most houses and buildings are constructed without great concern for the durability of their components. The only criterion is low first cost. As has already been noted, this viewpoint was made possible largely by the abundant, cheap fossil fuel energy resources that have shaped the energy monoculture of the current U.S. society. As long as such cheap fuels-and, concomitantly, cheap machines and building materials made possible by cheap energy-were available, why should builders and owners be concerned about fuel costs over the lifetime of a house? Why should they care about the possibility of replacing a cheap electric water heater every 5 or 6 years? As long as long-range availability of cheap fuels was assured, there has been no need for concern. Today, however, the picture has changed. The rosy days of cheap energy are over.

Thomas A. Robertson of the Energy Center at the University of Florida discovered just how quickly the attitudes toward life-cycle costing have changed in regard to solar energy. In the summer of 1973, Robertson conducted a survey of houses with solar water heaters in Miami, Florida. As has been noted, while thousands of solar water heaters were built in Florida in the earlier years of the century, the industry today has dwindled to a few companies building only a few heaters per year.

1 Even using a higher amortization rate than 6 percent, costs of solar heat are appreciably cheaper than those for electric heating today.