Although tremendous progress has been made in improving the conversion of sunlight into electricity with solar photovoltaic cells, their widespread adoption is primarily limited by high costs. This paper explores the use of the Internet as a catalyst for the diffusion of solar photovoltaic technology by reducing market risk. With market risk minimized by a database generated by a community of pledged consumers, solar cell companies would be motivated to construct a "Solar City Factory." Such a factory would produce solar panels that would enable systems costs to drop below US$1 per Watt and thus be less expensive than fossil fuels in providing bulk electricity. This price would have a positivespiral effect encouraging many consumers to switch to solar electricity and transition the global energy infrastructure to renewable solar energy.
The Solar City Factory
The need for reducing market risk
The Internet as catalyst
Solar photovoltaic, or PV, cells convert sunlight directly into electricity. With no additional technical improvements the cost of solar electricity could be competitive with conventional electricity if solar cells were produced on a large scale. Current manufacturing plants are about 100 times too small to capitalize on the economy of scale that is necessary to make solar cells pervasive. This article explores the possibility of a community of consumers using the Internet to provide the market security needed to catalyze the largescale manufacture of solar photovoltaic cells.
Solar photovoltaic energy conversion is a sustainable and environmentally friendly method of producing energy (Pearce, 2002). PV produces no atmospheric emissions or radioactive waste during use. Therefore, when it replaces fossil fuel energy production it curtails air pollution. PV electrical production also discharges no greenhouse gases, such as carbon dioxide, so it will help offset emissions that contribute to global climate destabilization. As a distributed electrical generation source, this technology acts as a network (much like the Internet), and unlike the conventional grid. It is therefore much less susceptible to largescale power outages caused by natural or manmade disasters. PV could also assist in national energy security and longterm economic growth for any country that aggressively develops the technology (d’Estaintot, 2000).
International cooperation and technology investment over the past 25 years has resulted in fantastic gains in solar PV cell performance. Solar cells made from a variety of materials have demonstrated efficiencies over ten percent and are currently manufactured globally. As the technological proficiency of the solar cell industry matured, the total shipments of solar cells increased rapidly. Currently, the U.S. PV industry has grown at average annualized rates of about 20 percent, a growth rate comparable to that of the semiconductor and computer industries (National Renewable Energy Laboratory, 2003). This growth rate, while impressive, must be kept in context of the global energy market. In 2000 the peak electrical generation capacity in the U.S. was 825 GigaWatts (GW = 109W) while the cumulative total global installed solar PV was less than a single GW. In the last four years the market has surged although it is still a tiny fraction of the overall global energy supply. The world PV installations in 2004 rose to 0.93 GW, representing growth of 62 percent over 2003 installations and the consolidated world production of PV increased to 1.15 GW (Solarbuzz, 2005a).
With no additional technical improvements the cost of solar electricity could be competitive with conventional electricity if solar cells were produced on a large scale.
As the production volume of solar cells increases the price per module falls rapidly, just as the price of computers plummeted from millions of dollars to only several hundred when in mass production. This trend is apparent for solar cells from 1975 to the present. The past two decades have seen a sustained price reduction of 7.5 percent per year during which the average worldwide production of modules increased by 18 percent per year (Shah, et al., 1999). The economic figure of merit for a solar cell is dollars per peak Watt (US$/Wp) . The price per Watt peak has fallen from over US$20.00/Wp to US$3.36/Wp as of June, 2005 (Solarbuzz, 2005b) . Unfortunately, the cost of solar electricity fully installed at even US$3/Wp is US$0.09/kWhr, which is still not competitive with the average cost of electricity in the U.S. of US$0.08/kWhr. Solar electricity will enjoy market dominance if the price of installed solar systems can be brought down to about US$1/Wp. At this price, the payback time from savings for a consumer’s electric bill is about five years in locations with high utility costs in the U.S. (e.g. US$0.12/ kWhr) and is much shorter in Japan and Germany with rates ~US$0.20 per kWhr. By reducing payback times the demand for solar PV panels on residential and commercial rooftops would be greatly increased.
The Solar City Factory
The ambitious target of US$1/Wp for installed PV can be reached by a massive scaleup of solar cell manufacturing facilities. Keshner and Arya (2004) propose the construction of a "Solar City Factory" — a factory capable of making 2.13.6 GW of solar cells per year. In this factory raw materials would be converted into the components of solar photovoltaic systems on site. Subfactories, housed within the City, will have equipment and processes that are sized, dedicated and optimized to produce only PV panels. The enormous size and integration of such a plant would enable a solar cell company to achieve new economies of scale and thus drop the price per peak Watt by a third to be economically competitive with fossil fuel fired electricity. The Solar City Factory will achieve such cost decreases by utilizing five key principles (Keshner and Arya, 2004):
- Using longlife substrate, packaging, and mounting materials. These materials must survive for 25 to 30 years on the outside of buildings, exposed to full sun, temperature and humidity cycles and in all climates of interest.
- Minimizing transportation and handling. Clean tunnels between equipment could be fully automated and used rather than standard clean rooms to eliminate handling within the factory to reduce cleaning, breakage and costs while improving yields.
- Dedicating and optimizing production of materials for solar panels (e.g. substrate glass specifically manufactured without iron for solar cells can increase the sunlight entering the cell by about 15 percent).
- Using multiple lines of identical equipment with rotating planned downtime for cleaning and maintenance. By minimizing unplanned downtime and having access to planned equipment maintenance the cost of the equipment, and the cost of maintenance and operations will decrease.
- Achieving high utilization rates for high purity input materials, which will both reduce costs directly and improve availability of rare materials for some types of solar cells (e.g. tellurium, selenium and indium).
The need for reducing market risk
A Solar City Factory would require a capital investment of around US$600 million (Keshner and Arya, 2004). Although this is a substantial amount it is a viable investment for a large company, when the market risk is low. Both semiconductor and flat panel display factories cost well over one billion dollars each. Historically, the solar industry has had a relatively high market risk because of volatile energy prices and widely swinging government support. The key to encouraging the major solar cell manufacturing companies to undertake financing a Solar City Factory is decreased market risk. A method of providing reduced risk must be employed to accelerate the construction of such a Solar City Factory.
A 100 MWp per year amorphous silicon fabrication plant is within reach of our current technology; unfortunately industry appears to be reluctant to take even such a small a risk.
In order to take advantages of economy of scale and to directly compete with fossil fuel as an energy source, it may not even be necessary to produce a Solar City Factory on the order of GWs. It is generally agreed that even a 100 MegaWatt peak (million peak Watts or MWp) thin film amorphous silicon PV plant would drop costs far enough to compete directly with conventional electricity in some gridtied markets (Schramm and Kern, 2000; DeMeo, 1999; and, Payne, et al., 2000). For PV amorphous silicon modules produced at a 100 MWp plant once financing, capital equipment costs, direct and indirect manufacturing costs, installation, power conditioning, operation and maintenance costs, and tax benefits are all taken into account, the installed PV price is likely to fall under US$3.00/Wp(AC) (Payne, et al., 2001). Payne, et al. report that this price makes solar cells economically attractive to a substantial portion (~8,000 MWp) of the United States’ domestic energy market. The enormous market generated from the first amorphous silicon 100 MWp/yr plant has obvious implications for energy policy. In addition, any increases in electrical prices would further accelerate the market. A 100 MWp per year amorphous silicon fabrication plant is within reach of our current technology; unfortunately industry appears to be reluctant to take even such a small a risk. This underscores the necessity of identifying a secure market, so such a plant would be quickly constructed.
The Internet as catalyst
The Internet could be used to provide the low risk market necessary for investment in largescale solar cell production. The plan is very simple and can easily be instituted with current Internet technology. Energy consumers would sign up on a Web site by giving their names, addresses, phone numbers, email addresses, and "pledge levels." Your pledge level would indicate how many peak Watts of solar panels you would be willing to purchase based on the price of an installed system. Your identifying information would be kept confidential; however, your participation level would be posted on a publicly accessible Web site. This information could be displayed visually as seen in Figure 1. It would show in real time how many people agreed to a certain level of participation. In this way consumers become "subcontractors" to the major solar cell manufacturers providing needed market data as envisioned by Dolfsma (1999). The Internet would make possible the realtime recording of market information from selfselected future solar cell consumers.
Figure 1: A simulated example of the pledged purchases of solar PV [Wp] on a log scale as a function of the price of PV [US$/Wp] for an installed solar PV system. As the price of solar cells decreases linearly the demand for solar cells increases by orders of magnitude.
This plan could be used by individual companies, for example, General Electric, or the solar industry as a whole through the Solar Energy Industries Association of Solar Energy International. Similarly, an interested government could encourage citizens to pledge by offering modest tax deductions, rebates, or encouraging the pledging into other online services (e.g. change of address). For some companies or institutions this could arguably create the entire demand, e.g., if there was a focused effort for parishes in the U.S. to place a PV array on the roof of their churches over half the demand could be met by the U.S. Catholic Church alone . In addition, with net metering laws  taking effect in a growing number of states, PV system owners can sell any electricity they do not use back to electrical providers, and further increase demand.
The Internet could be used to provide the low risk market necessary for investment in largescale solar cell production.
To encourage consumers/subcontractors to pledge they would have to be compensated. If the Internet market risk reduction program were being run by an umbrella group, consumers could receive email updates of current market prices, be entered in drawings, be given coupons or directly compensated. There are already numerous examples of direct consumer payments for information on the Internet (Henshall, 2000). The Web sites www.surveyplatinum.com, www.surveyscout.com, www.surveyclub.com, and www.freeride.com are examples of how consumers are being paid by data aggregators to gain marketing information. However, the incentives most likely to garner reliable and widespread participation would be discounts from a specific manufacturers or the guaranteeing the right to purchase solar panels at a given cost before they are made available to the public. It is likely that if solar electricity were competitive or lower costs than conventional electricity supply, even with a 3.6GW plant in full production would not be able to meet the demand. In this case waiting lists would be necessary similar to the current situation with Toyota’s hybrid vehicles.
In order for this plan to succeed the solar PV manufactures would need to have a firm commitment from the consumers, which is somewhat different from the assurance provided by an online survey. The purchase pledge could be viewed as a formal buyer’s agreement, which if properly designed would be interpreted by the courts to be a legally binding contract. A similar legal situation already exists for sponsors of contests with the contractual obligation to pay the winners according to their public announcement (Sullivan, 1988). A contestant who performs the requested act has fulfilled the requirements for a legally binding contract with the sponsor. Attempts by a contest sponsor to renege by changing contest rules after a contestant has accepted the offer by performing the desired act are generally treated as breach of contract (Kremer, 2001). Advertisements with quantified specifications (e.g. definite quantity of a product) for the purchase of products at given prices have also been found to be legally binding (Vaccaro, 1972). In this case, each individual consumer could be viewed as a sponsor of a contest in which the ‘prize’ is the sponsor purchasing a specific number of pledged Wp of solar panels. The ‘act’ would be providing the solar panels at the requested purchase price in dollars per Wp. There are several examples of such contestmodeled programs to encourage the development and mass deployment of new technologies: energy efficient refrigerators (Feist, et al., 1994), human powered flight (Grosser, 1991), vaccines (Kremer, 2001), or civilian space flight via the Ansari X Prize .
The COMsumer  Manifesto states, "The consumers’ position in the world is changing. Through technology the Internet is framing a new revolution, changing the underlying ecology and enabling a new form of empowered consumer to emerge." (Henshall, 2000). Henshall believes that information belonging to communities of consumers will be the most important resource, in the new knowledge economy. This paper offers another example to support this idea of the possibility of the realworld market and manufacturing being directly guided by an exchange between communities of consumers empowered by the Internet.
Using the Internet to first catalyze the largescale production of a socially supported technology such as PV is an optimal method of formalizing the role of COMsumers.
Using the Internet to first catalyze the largescale production of a socially supported technology such as PV is an optimal method of formalizing the role of COMsumers. The vast majority of the global public is in favor of solar PV because of its numerous positive attributes most notably those revolving around environmental stewardship and energy independence. For example, the Program on International Policy Attitudes found that the American public wants the federal budget for renewable energy research like solar PV to increase by 1090 percent (Kull, et al., 2005). It is largely because of such positive attitude towards PV that consumers would be likely to participate in the numbers necessary to guarantee the market for a Solar City Factory. If we take the scenario of a 3.6GWp factory to break the US$1/Wp barrier and assume the average house would need 5,000Wp to provide for its energy needs, only 720,000 consumers would need to individually pledge. This represents only about two percent of the population of California and a truly miniscule fraction of the global population.
Finally, it should be noted that utilizing the Internet as a tool for the diffusion of innovation in this way could also be useful for other ecologically beneficial technologies that demand a large investment such as wind farms, fuel cell vehicles, and the infrastructure for a hydrogen economy.
The Internet could act as a catalyst for the diffusion of solar photovoltaic technology by decreasing market risk. With a market risk approaching zero provided by an Internet database generated by a community of pledged consumers, solar cell companies would be motivated to construct a 3.6 GW Solar City Factory. This Solar City Factory would produce solar panels that would enable systems costs to drop below US$1/Wp. This price would have a positivespiral effect encouraging many consumers to switch to solar electricity and transition the global energy infrastructure to renewable energy.
About the author
Joshua M. Pearce is an assistant professor of physics at the Clarion University of Pennsylvania where he researches solar photovoltaics and applied sustainability.
Email: jpearce [at] clarion [dot] edu
1. Watt peak is the power output in Watts of a solar cell module when it is illuminated under standard conditions of 1,000 Watts/meter² intensity, 25oC ambient temperature and a spectrum that relates to sunlight that has passed through the atmosphere (AM or Air Mass 1.5). This would be equivalent to placing your solar cell outside pointed directly at the sun on a clear day.
2. It should be noted that this is the price of the panels alone and does not include installation or the balance of systems costs, which include mounting, wiring, inverter and in some cases charge controller and batteries.
3. In order to obtain a rough estimate of the electrical generating capacity of the entire continental U.S. Catholic fleet of churches, I utilized a Model Catholic Church or MCC. Following standard rulesofthumb for church sizing requirements, and assuming an MCC with a 45o roof, and an average seating capacity of 400, the total roof area is ~1,144 m². Only half of total roof area is utilized (the half facing the most sun or the south in the U.S.) with PV. So the area available for a PV array is about 570 m² per church. To cover half the roof of an MCC with a modern HIT solar cells with power densities of about 141.5 W/m² would require ~80kW of panels. There are 20,842 parishes in the continental U.S. (Cheney, 2005). Assuming they are MCC and covered with such panels, they would place a demand of over 1.68GW on the solar industry.
4. Net metering works as follows: The electricity (measured in kWhrs) delivered to an electrical provider from a PV system are subtracted from the kWhrs delivered from the electrical provider for each billing cycle. If the kWhr calculation is net positive for the billing cycle, the electric provider will bill the net kWhr to the customer under the applicable price plan. If the kWhrs calculation is net negative for the billing cycle, the electric provider will credit the net kWhrs from the customer at an average market price.
5. See http://www.xprize.org/home.php and http://www.xprizefoundation.com/ for details.
6. COMsumer, the word coined to describe communities of consumers on the Internet comes from the Latin com plus sumere meaning to take together (Henshall, 2000).
David M. Cheney, 2005. "Catholic Hierarchy, Statistics by Province," at http://www.catholic-hierarchy.org/country/spcus1.html, accessed 12 May 2004.
T. d’Estaintot, 2000. "European Commissionsupported R&D activities in the field of photovoltaics," In: Proceedings of the 28th IEEE Photovoltaic Specialists Conference (1522 September, Anchorage), pp. 17341735.
Edgar DeMeo, 1999. PV: Cost Reduction Prospects, Renewable Energy Consulting Services, Inc. for the International Finance Corporation, 30 March 1999.
Wilfred Dolfsma, 1999. "Consumers as Subcontractors on Electronic Markets," First Monday, volume 4, number 3 (March), at http://www.firstmonday.org/issues/issue4_3/dolfsma/, accessed 12 June 2005.
John Feist, Ray Farhang, Janis Erickson, Elias Stergakos, Paul Brodie, Paul Liepe, 1994. "Superefficient refrigerators: The golden carrot from concept to reality." In: Proceedings of the ACEEE 1994 Summer Study on Energy Efficiency in Buildings, volume 3. Washington, D.C.: American Council for an Energy Efficient Economy, pp. 6776.
Morton Grosser, 1991. Gossamer odyssey: The triumph of humanpowered flight. New York: Dover.
Stuart Henshall, 2000. "The COMsumer Manifesto: Empowering communities of consumers through the Internet," First Monday, volume 5, number 5 (May) at http://firstmonday.org/issues/issue5_5/henshall/, accessed 12 June 2005.
M.S. Keshner and R. Arya, 2004. "Study of potential cost reductions resulting from superlargescale manufacturing of PV modules," National Renewable Energy Laboratory, Final Subcontract Report NREL/SR52036846, at http://www.nrel.gov/ncpv/thin_film/docs/keshner.pdf, accessed 12 June 2005.
M. Kremer, 2001. "Creating markets for new vaccines: Part II: Design issues," In: Adam B. Jaffe, Josh Lerner, and Scott Stern (editors). Innovation policy and the economy. Volume 1. Cambridge, Mass.: MIT Press; see also http://ideas.repec.org/p/nbr/nberwo/7717.html, accessed 29 July 2005.
S. Kull, C. Ramsay, S. Subias, S. Weber, and E. Lewis, 2005. "The Federal budget: The public’s priorities," PIPA/Knowledge Networks Poll (7 March), at http://pipa.org/OnlineReports/budget/030705/Report03_07_05.pdf, accessed 12 June, 2005.
National Renewable Energy Laboratory, 2003. "Solar electric power: The U.S. photovoltaic industry roadmap," at http://www.nrel.gov/ncpv/pdfs/30150.pdf, accessed 12 June 2005.
Adam Payne, Richard Duke, and Robert Williams, 2001. "Accelerating PV expansion: Supply analysis for competitive electricity markets," Energy Policy, volume 29, number 10, pp. 787800.http://dx.doi.org/10.1016/S0301-4215(01)00014-3
Adam Payne, Richard Duke, and Robert Williams, 2000. "The impact of Net metering on the residential rooftop PV market," In: Proceedings of the 28th IEEE Photovoltaic Specialists Conference (1522 September, Anchorage), pp. 13911394.
Joshua Pearce, 2002. "Photovoltaics A Path to Sustainable Futures," Futures, volume 34, number 7, pp. 663674.http://dx.doi.org/10.1016/S0016-3287(02)00008-3
G. Schramm and E. Kern, 2000. "Accelerating photovoltaic production through grid connected applications in developing countries," In: Proceedings of the 28th IEEE Photovoltaic Specialists Conference (1522 September, Anchorage), pp. 3639.
A. Shah, P. Torres, R. Tscharner, N. Wyrsch, and H. Keppner, 1999. "Photovoltaic technology: The case for thinfilm solar cells," Science, volume 285, number 5427, pp. 692698.http://dx.doi.org/10.1126/science.285.5428.692
Solarbuzz, 2005a. "Marketbuzz 2005: Annual world solar photovoltaic (PV) market report," at http://www.solarbuzz.com/Marketbuzz2005-intro.htm, accessed 12 June 2005.
Solarbuzz, 2005b. "Price survey June 2005: Lowest thin film module price," at http://www.solarbuzz.com/index.asp, accessed 12 June 2005.
Michael P. Sullivan, 1988. "Private contests and lotteries: Entrants’ rights and remedies," American Law Reports, 64 A.L.R.4th.
Don F. Vaccaro, 1972. "Advertisement addressed to public relating to sale or purchase of goods at specified price as an offer the acceptance of which will consummate a contract," American Law Reports, 43 A.L.R. 3rd.
Paper received 8 July 2005; revised 19 July 2005; accepted 21 July 2005.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 2.5 License.
Using the Internet to reduce market risk for alternative energy sources: The case of largescale solar photovoltaic production by Joshua M. Pearce
First Monday, volume 10, number 8 (August 2005),