There are two possible sources of carbon-free power available today: solar power and nuclear power.  Solar power includes solar photovoltaic (converting sunlight directly into electrical power) and solar thermal (heating a fluid with sunlight and converting the heat to electricity).  Solar power has received hundreds of billions of dollars in subsidies and research grants in the United States alone, and perhaps as much as a trillion worldwide over the past 30 years.  For that investment, solar contributes roughly 1% of the world's total power.

So which is better, solar or nuclear?  While it is impossible to provide a definitive answer now, there is a gigantic laboratory in Europe that gives a strong indication.  That laboratory is France and Germany.  For years, France has embraced nuclear power, and currently, it gets about 75–80% of its electric power from its 58 nuclear reactors.  Germany has embraced an "energiewende," or energy transformation.  It intends to get as much power as possible from solar, while disassembling its nuclear infrastructure.   Currently, it is getting 25–30% from solar.

Let's compare France and Germany on price for electricity and emission of CO2 into the atmosphere.  Despite massive German subsidies, the price for power in Germany is rising fast, and it is approaching about 40 cents per kilowatt-hour.  In France, the price is less than half of this (in the United States, less than a third).  Yet this is for solar producing less than a third of German electric power, whereas nuclear produces nearly all of the electric power in France.

Regarding per capita CO2 injection into the atmosphere, the French inject about half of what the Germans do.  Both countries have cars and trucks that use gasoline; however, in the electrical generating sector, the French emit little CO2, while Germany emits a great deal.  It still must generate around two thirds of its power in the conventional way and must keep conventional thermal generators going for when the sun does not shine or the wind does not blow.  At this point, at least, honors in the competition go to France and nuclear power. 

To see the dilemma facing solar power, it is necessary to go through some numbers; these bore people, but they are important.  A typical coal, gas, or nuclear power plant generates about a billion Watts, or 1 gigawatt (1 GW), and occupies about one quarter of a square kilometer.

At high noon on a summer day, the total solar power is about 1 GW per square kilometer.  Averaging over solar angle, day and night, rain and shine, this is about 1 GW in about two square miles.  But sunlight is converted to electricity by solar panels with only about 10–20% efficiency, so a 1-GW solar panel power plant would occupy 10–20 square miles of land that could not be used for anything else.  Perhaps this would work in Nevada or Arizona, but in the Northeast, Pacific Coast, or Midwest, that sort of land is simply not available, for a 1-GW plant, at a reasonable price.  Even in Nevada or Arizona, there are occasional dust storms, and the solar panels have to be kept reasonably clean in a land with little water.

Wind power is even more daunting.  A 1-GW wind farm would occupy about 190 square miles.  At least crops can grow, and animals can graze, but this area is totally unfit for humans.  Not only are the noise and vibrations intolerable, but in winter, blades could be ice-covered, and occasionally chunks of ice, hundreds of pounds, could be thrown hundreds of yards.  Any person hit by one would certainly be killed.

The lesson is clear.  Sunlight is free, but solar-generated electricity is very, very expensive. 

Wallace Manheimer is a life fellow of the American Physical Society (APS) and the Institute of Electrical and Electronic Engineers (IEEE).  His career has been at the U.S. Naval Research Laboratory since 1970, and he served in the small group of ST-16 senior scientists for his last 14 years there.  Since retiring in 2004, he has served as a consultant at the lab.  At NRL he has worked on inertial fusion, magnetic fusion, a nuclear disturbed upper atmosphere, electron and ion beams, high power microwave and millimeter wave systems, advanced radar systems, and plasma processing.  He is the author of over 150 refereed scientific papers.