Does Solar Energy Have a Future in Utility Power?
If so, When and Why?
George D. Cody
Scientific Advisor, Exxon Corporate Research
About the Lecture
Photovoltaics (PVs) generate a significant fraction of the power in underdeveloped countries and supply the power of earth-orbiting satellites. Why aren't we using more solar energy instead of fossil fuels? Can PV power, which currently costs more than five times the US average, ever be competitive? And if so, over what time period can this be achieved?
To answer these questions, we use a utility pricing formula and experience curves to project the cost and performance of PV power over the next 15 years suggesting parity with utility average power by the year 2010 in selected locations. Recent press releases of Amoco/Solarex/Enron projecting sales of PV power at rates below the current US utility average starting in 1999 (!) in Nevada are placed in the context of our early projections. Finally we discuss the opportunities and limitations on the conversion of sunlight to CO2-neutral biomass fuel as well as fundamental solar limits on world food supply.
About the Speaker
George D. Cody received AB (1952) and PhD (1957) degrees from Harvard University. Mr. Cody was a staff member and laboratory director at RCA Laboratories (1957-76) and scientific advisor at Exxon's corporate research laboratory (1976-98). He received the Franklin Institute's Ballantine medal in 1979 for the invention of the Ge-Si thermoelectric that powers the Voyager, Galileo, and Cassini spacecrafts. His recent research focussed on the optics of amorphous semiconductors, and non-intrusive probes of two-phase flow in refining and petrochemicals. He has published more than 100 articles and holds 13 patents. He is a fellow of the American Physical Society and a member of the American Institute of Chemical Engineers.
President Garavelli called the 2104th meeting to order at 8:20 p.m. on April 2, 1999. The Recording Secretary read the minutes of the 2103rd meeting and they were approved.
The speaker for the 2104th meeting was George D. Cody of Exxon Corporate Research. The title of his presentation was “Does Solar Research Have a Future in Utility Power? If So, When and Why?”
As an interesting aside to Mr. Cody's talk, his cousin and our fellow PSW member, Martin Cody, introduced him. It turns out that the two cousins lost touch with each other more than 50 years ago and ironically crossed paths here at the PSW. Thus, the 2104th PSW meeting not only looked to technological aspects for the future, but also reinforced personnel roots in the past.
From about the middle of the last century to the present, fuel supplies for the United States changed about every 70 years. For example, in the 1840's, wood made up about 90% of the fuel supply; 70 years later, about 1910, coal made up about 80% of the fuel supply and then 60 years later, in the 1970's, oil and gas comprised 80% of the supply. What lies ahead for our source of fuel? More specifically what lies ahead for solar energy and photovoltaics?
Mr. Cody pointed out that photovoltaics already generate a significant fraction of the power in underdeveloped countries and supplies the power of earth-orbiting satellites. Why then aren't we using more solar energy instead of fossil fuels? Can photovoltaic power, which currently costs more than five times the US average, ever be competitive? And if so, over what time period can this be achieved?
To begin answering these questions the speaker pointed out that in 1988 photovoltaic systems had more than 100 times the efficiency in converting sunlight to energy than did trees. Consequently, photovoltaic authorities predicted that it would be a gross underestimate of the competence of our scientific community and of the ingenuity of our industry not to believe that they could design a solar cell system significantly cheaper than nature's forests given a hundred-fold advantage in efficiency.
With that as a motivating thought for photovoltaic developers, Mr. Cody went on to describe a utility pricing formula and experience curves to project the cost and performance of photovoltaic power over the next 15 years. The results suggest parity with utility average power by the year 2010 in selected locations. Further, recent press releases of Amoco, Solarex and Enron already project sales of photovoltaic power at rates below the current US utility average starting in 1999 in Nevada, which fit in the context of early projections.
In the way of background for pricing formulas and experience curves, the speaker described extensive studies of manufacturing companies over the last 40 years which led to the observation that as organizations produce more of a product, the unit cost of the production typically decreases at a decreasing rate. This phenomenon is referred to as a “learning curve, a progress curve, an experience curve or learning by doing.” The learning curve is a generic description of the observed empirical relation between the cost of a product and the cumulative quantity of the product. In a large number of cases the functional relation between the two quantities shows that costs often go down linearly on a log-log scale. A power law can thus express this relationship and the exponent is defined as the learning parameter.
Mr. Cody advised that in 1976 photovoltaic cost about $5.00 kW-hr and in 1988 the cost dropped to $0.80 kW-hr. Now applying this experience and the empirical learning curve approach to the resulting projections put the cost of photovoltaics at $0.04 kWh in 2006.
Mr. Cody then closed his presentation and kindly answered questions from the floor. President Garavelli thereupon thanked Ms. Cleave for the society, announced the next meeting and made the usual parking announcement. He then adjourned the 2104th meeting to the Social Hour at 9:40 p.m.