4 The future of direct solar energy use
The immense global energy flux from the Sun makes it the prime candidate for future sustainable energy production. Both solar thermal energy and solar PV conversion involve technologies that can be deployed on personal through community to regional scales, using both simple and advanced technologies. You have probably seen solar PV panels that power automatic roadside weather stations and other low-drain communications systems. The panels require low maintenance and usually charge batteries to allow them to remain operational during the night. In poor countries where the energy infrastructure is rudimentary or absent, PV systems hold out great potential. An important use is for daytime pumping of water from wells.
In its 1997 renewable energy plan, the EU set a target of half a million villagescale direct solar systems to be deployed in developing countries and a similar number in European houses by 2010. The United Nations has asked world governments to deploy 4.5 GW of solar PV electricity generating capacity in developing countries by 2012. Both Japanese and German governmental subsidies have boosted both photovoltaic production and deployment, and a similar 'hard-sell' stems from commercial sources in the US. As a result, power capacity of solar PV rose from a global 50 MW in 1995 to over 2 GW in 2002, and is estimated to be growing at a rate of 40% per year. The main hindrance to greater deployment is simply that of cost; at between US$ 0.2 to 0.5 per kWh, solar PV electricity was almost ten times as expensive in 2005 as that from the cheapest fossil-fuel source, natural gas. To progress, the technology requires continued reduction in the cost of the solar cells themselves - but the enormous reduction in cost of silicon-based computer hardware since the 1970s is cause for optimism.
Solar PV could theoretically supplement grid-power during daylight hours to reduce generating costs and environmental emissions. However, at this scale serious disadvantages emerge. The daily intensity of sunlight varies dramatically because of cloud cover. Moreover, solar power is greater in the summer while the demand for electricity is lower, except in areas with high use of air conditioners. Provided photovoltaic conversion contributes no more than 10-20% of the total amount of electricity in the grid, its integration seems feasible. This is because electricity grid systems are designed to cope with large variations in demand, and they can cope equally well with fluctuations from different forms of supply.
Should future solar PV power rise above 20% of the total electricity supply, then existing grid systems built to be dominated by coal, oil and nuclear generation would have to be modified. This is because conventional power plants are slow to start up and shut down; they are slow-response systems. Solar conversion, along with other alternative sources whose power source fluctuates uncontrollably (e.g. wind and waves), is a fast-response system. A distribution grid with solar PV power as a major component would need to be supplemented by controllable fast response systems, principally hydroelectric and gas-turbine generators. Another solution would be short-term electrical storage installations, but they are both costly and inefficient. A means of 'having one's green cake and eating it', however, would be to use electricity from solar PV to generate hydrogen by electrolysis of water. Hydrogen gas is combustible, storable and moveable, and so avoids most of the problems associated with electrical storage. The hydrogen could even be converted back into electricity using fuel cells.
Despite these caveats, the potential of solar PV is enormous. If photovoltaic conversion with 10% efficiency was installed over an area of 500 000 km2 (about 1.3% of the area of tropical deserts) humanity's present energy requirements would be met. That outlook is probably far off. Of the electricity generated from all alternative energy sources in the early 21st century, solar PV contributes only about 0.02%, with solar thermal generation a little more significant at 0.06%.