Physicists define energy as "the conserved quantity associated with the temporal invariance of the Lagrangian density." This rigorous and fully general definition is probably incomprehensible to anyone but a physicist, so we often use the more informal definition of energy as "the capacity to do work." Physicists measure energy in joules, chemists traditionally use kilocalories, and the electric power industry prefers kilowatt-hours. It takes about twelve joules to raise the temperature of a teaspoon of water by one degree Fahrenheit. A kilowatt-hour is 3,600,000 joules, while a kilocalorie is 4186.8 joules. The "calorie" familiar to dieters is actually a kilocalorie.
Power is defined as the rate at which energy is supplied. The power rating of engines and motors is measured in horsepower, while the power rating of almost everything else is measured in watts. A watt is one joule per second, and a horspower is about 745.7 watts. Ironically, the watt is named for James Watt, who measured power in units of horsepower. It has been observed that one horsepower corresponds to a rather sturdy horse.
In pre-industrial societies, mechanical work was supplied by the muscles of people and their domesticated animals. Heat for carrying out manufacturing processes, such as cooking or metalsmithing, or simply for keeping a house warm, was supplied by burning wood or other biomass. At night, light was supplied by burning animal fat or vegetable oil, assuming it was supplied at all. In every case, the ultimate source of energy was nuclear fusion in the heart of the Sun. This energy reached the Earth as sunlight, was absorbed by green plants, and converted to chemical energy in the form of starch, cellulose, or other edible or combustible compounds.
With the industrial revolution, the chemical energy of fuels was more effectively harnessed by converting it to mechanical work in a steam engine. This also allowed a wider range of fuels to be used, since steam engines are less discriminating about what they consume than humans or animals. (It is true that I have seen pigs eat coal, and I have never seen a steam engine burn oysters, but the general observation still applies.) In fact, almost any heat source can be used in a steam engine, including solar power and nuclear power.
The final step in the industrial revolution was the invention of the dynamo by Michael Faraday. This is a device that converts mechanical work into electrical energy. Electrical energy is one of the most versatile forms of energy, easily converted back to mechanical energy, to heat, or to light. It can also be used directly in industrial electrolysis processes, such as the Hall process for smelting aluminum.
Most electrical energy is generated today using steam turbine generators. This is a combination of the steam engine and the dynamo. Heat energy is used to produce high-temperature steam, which is used to drive a special type of steam engine called a steam turbine. (This converts the heat energy into mechanical work.) The mechanical work from the steam turbine is then used to drive a dynamo, which converts it into electrical energy that can then be fed into the power grid.
The steam turbine generator has been around for a long time, is well understood, and can be designed to operate close to its maximum theoretical efficiency. This theoretical efficiency is determined solely by the temperatures of the hot steam and of the surrounding environment, and it cannot be improved further by any conceivable advance in technology. The only thing that distinguishes the steam turbine generator in a nuclear plant from one in a coal-fired or solar plant is the original source of the heat and the design of the boiler used to generate the hot steam.
In a fuel-burning power plant, coal or another fuel is burned to yield heat, which is applied directly to the coils of a boiler. A significant amount of heat literally goes up the chimney, along with large quantities of pollutants such as particulates, sulfur oxides, heavy metals, nitrous oxides, and carbon dioxide. Other noncombustible contaminants in the fuel form a solid clinker that must be disposed of in a landfill. Each kilowatt-hour generated by a coal-fired plant requires the combustion of about a third of a pound of coal, which produces one or two ounces of clinker and other pollutants.
In a nuclear plant, the source of heat is the fission of atomic nuclei of heavy elements. Unlike chemical reactions, which are governed by the electromagnetic force, nuclear reactions are governed by the strong force, which is about a million times as strong as the electromagnetic force at short distances. As a result, the reactions are about a million times as energetic. A pound of uranium can theoretically produce as much energy as almost 2000 tons of coal. This also means that a coal-burning plant produces several hundred tons of waste for each pound of waste from a nuclear reactor. However, while both kinds of waste are toxic, the nuclear waste is very much more toxic (on a pound-for-pound basis) than the clinker.
Thus one advantage of a nuclear power plant over a coal-fired power plant is that the nuclear plant requires much less fuel and produces a much smaller amount of (admittedly much more toxic) waste. But the more important advantage is that known reserves of coal in the United States could supply our total energy needs for perhaps another sixty years, while known nuclear fuel reserves are theoretically good for at least another 400 years. After that time, we must hope that our scientists will have discovered how to control nuclear fusion, which uses a hydrogen isotope found in seawater. If fusion can be made practical, then the amount of fuel in seawater is so vast that we cannot foresee a time when our fuel supply will run out.
Obviously, there are some challenges with nuclear power, or we would be using it everywhere already and we would not have very dedicated activists opposing any further nuclear power plant construction. Three are most important.
First, there is the problem of disposing of the highly dangerous waste from nuclear fission. I have discussed this here, and believe it is a readily manageable problem from a technical point of view. The French, who currently get a very large fraction of their electrical power from nuclear plants, have developed several promising innovations (such as borosilicate glass vitrification) that reduce the risks of waste disposal. There are obvious political difficulties with determining the location of the waste disposal plant that can be overcome only with sufficient grass-roots pressure.
Second, there is the problem of nuclear proliferation. Ordinary reactor fuel is uranium-238 mixed with a small amount of uranium-235, with the uranium-235 supplying the energy. It is impossible to make nuclear explosives from this mixture of isotopes, and separating the isotopes is sufficiently difficult and costly that efforts to do so are difficult to hide. But using only this fuel mixture is wasteful, because the U-238 remains unused. A breeder reactor converts the U-238 into plutonium, which can then be fissioned to yield energy. However, the plutonium so produced can be used in nuclear explosives without the costly isotope separation process. This has led to opposition in arms control circles towards breeder reactors. Ultimately, this is a political rather than technical question, but my opinion is that a breeder reactor under U.S. control is a negligible proliferation risk compared with nuclear technology that is already in the hands of proliferant states. Can you say "Iran?" Or "North Korea?" I knew you could.
Third, there is the danger of a Chernobyl-style reactor accident with widespread environmental damage. I believe the risks here are greatly overstated for two reasons: First, the health danger from radiation is routinely overestimated due to the use of a linear no-threshold model that is grossly pessimistic. Second, the Chernobyl reactor was horribly designed by Western standards. Our own worst accident, at Three Mile Island, resulted in a negligible release of radiation in spite of the fact that the reactor crew did almost everything wrong and managed to melt half the reactor core. We have subsequently made advances in reactor design that eliminate most of the risk of even another TMI-style accident.
The TMI reactor, like most reactors built in the last century in this country, relied on engineered safety to prevent an accident. The reactor was controlled using long rods containing elements that soak up the neutrons necessary for maintaining the nuclear reaction. These could be inserted deeper into the nuclear pile if the reaction became too vigorous. Such a system is known as an actively controlled system. It is vulnerable to various kinds of accidents if the active control system fails, as apparently happened at TMI. So the reactor core is surrounded by multiple layers of steel, concrete, and other materials meant to contain the radioactive material in the core if an accident does occur. These multiple layers pretty much did their job at TMI, though they did not prevent a half-billion-dollar reactor core from destroying itself.
Newer reactor designs use passive control, in which the design of the reactor is such that any increase in the reaction rate automatically increases the loss of neutrons without any active intervention. Such reactors are described as intrinsically safe. A friend of mine, a fellow named Ray, was working on such a reactor back during the Carter administration. The funds for the project were cut the same day they first achieved reactor criticality, which is another good reason to despise "history's greatest monster."1
There is one other disadvantage of nuclear power: Nuclear plants are quite expensive to build. Much of this cost is in the licensing process. It is likely that this cost could be amortized through standardization of design, as Canada has done with the CANDU reactor. In any case, the cost of conventional fuels can only go up, and it is inevitable that nuclear power will someday be competitive. Estimates of when that day will come range from "it has already" to "not for another fifty years." Since the more pessimistic estimate comes from Steve Verdon, who is an economist with connections to the energy industry, I am not unduly optimistic. But my opinion is that resumption of the construction of nuclear power plants is inevitable, and sooner is better than later.
1This bit of hyperbole comes from The Simpsons.