All large power stations operate on a similar principle - they consist of a heat source that is used to produce steam, which then turns a turbine connected to an electrical generator. The main differences lie in the type of heat source. The heat can come from the combustion of fossil fuels such as coal and oil, or from a nuclear reactor. Normal combustion processes release energy as a result of the making and breaking of chemical bonds, whereas in a nuclear reactor the energy comes from the fission of atomic nuclei. The amount of energy (i.e. heat) released in a single nuclear fission reaction is much greater than in the breaking of a single chemical bond, and the energy locked up in a gram of nuclear fuel is consequently much greater than in a gram of fossil fuel. Operating a 1 000-MWe power plant for one year would consume about two million tonnes of coal (and emit 3-4 times this amount of greenhouse gases, i.e. CO2), but only 25 tonnes of nuclear fuel (and no CO2 emissions).
The components of a nuclear reactor
The principal components of a nuclear reactor are the nuclear fuel, the moderator, the control rods and the coolant. There are many possible nuclear reactor sizes and configurations, with different fuel compositions, moderators, coolants and control mechanisms, all capable of allowing a nuclear chain reaction to be sustained and controlled. Many of these configurations have been investigated and studied in research programmes or as prototype commercial plants, though only a few have been developed to the point of industrial maturity. Economics are one of the drivers for this process, as are safety considerations - for example, some designs might be very efficient at producing heat, but have unfavourable characteristics that make them difficult to control in certain situations. The most successful reactor type is the Light Water Reactor (LWR), which uses ordinary water under high pressure as both the coolant and the moderator. LWRs come in two design configurations - the Pressurised Water Reactor (PWR) and the Boiling Water Reactor (BWR).
In LWRs, the water coolant needs to be kept under high pressure, so the reactor itself is a pressure vessel containing the fuel and control rods. In PWRs, the pressurised water circulates through the reactor vessel and round the primary circuit, where it transfers its heat to a secondary lower pressure circuit in the steam generators. The steam then drives the turbines which power the generator. In BWRs the cooling water is at a lower pressure than in PWRs and there is no secondary steam circuit - after passing through the reactor the cooling water is allowed to boil, with the steam then being separated off to drive the turbine. In both cases, the reactor vessel and primary cooling circuit are housed in a containment building made of pre-stressed concrete, which provides physical protection of the reactor, even from extreme events such as aircraft impact, and prevents any release or radioactivity to the environment even in the case of the most severe malfunction of the reactor or cooling system. The containment building is the most recognisable external indication of the reactor.
The LWRs currently in operation around the world, along with one or two other reactor types (such at the CANDU reactors or the AGR in the UK) are often referred to as 'Generation-II' reactors. Most reactors currently under construction are evolutionary versions of these designs that offer improved economics and efficiency, and are referred to as 'Generation-III' reactors. Scientists and engineers around the world are currently investigating advanced, so-called Generation-IV, reactor designs, which use very different basic components - fuel, coolant, etc. - and core configurations, can be operated in different ways and can even be used not only to produce electricity but also high temperature heat for industrial processes, thereby offering an alternative to burning of fossil fuels also in sectors other than power generation.
The fuel used in nuclear reactors contains enough fissile material (see 'a small dose of nuclear physics') to sustain a fission chain reaction over many months, if not years. Only the U-235 nuclei in uranium are fissile, but these nuclei make up only 0.71% of uranium found in nature. The other 99.3% of natural uranium consists of U-238, i.e. nuclei with 92 protons but 146 neutrons. The U-238 is not fissile, but it is 'fertile', since it will readily absorb a neutron and, instead of fissioning, transmute via a couple of radioactive decays into Pu-239, which is a fissile isotope of plutonium.
Typically, nuclear fuel is loaded into a reactor in the form of fuel elements. These are usually bundles of narrow metal tubes - called fuel rods or pins - held together in an array. The rods contain ceramic fuel pellets made of uranium oxide. Though it is possible to construct a reactor that can sustain a chain reaction using fuel containing only natural uranium, most reactor designs in use today require the uranium first to be 'enriched', through a highly complex industrial process, in order to increase the percentage of U-235 (typically up to 3-4%). In some reactors, plutonium oxide (i.e. containing predominantly the fissile isotope Pu-239) is also added. This plutonium comes either from recycled (reprocessed) fuel that has already spent time in a reactor or occasionally from warheads of decommissioned nuclear weapons.
The energy locked up in nuclear fuel and that can be released through fission of the U-235 or Pu-239 is millions of times higher than that contained in a similar amount of fossil fuel. Nonetheless, most of the uranium going into present-day reactors comes out unused. The fuel usually spends a few years in the reactor, and during this time the majority (but not all) of the U-235 undergoes fission (i.e. is 'burnt'). Crucially, the fuel will, after this period, no longer contain enough fissile material to sustain a chain reaction and must be replaced with fresh fuel. Most of the U-238 will come out unaffected, though a small but significant amount will have been 'transmuted' into Pu-239. In addition, small quantities of other heavy isotopes (actinides) will be created as a result of capture of neutrons, without fission, by the uranium and plutonium nuclei. These are collectively referred to as the minor actinides, and include isotopes of elements such as neptunium, americium and curium. The 'spent fuel' contains the minor actinides as well as the lighter fission products resulting from the fission of U-235 and Pu-239. These minor actinides and the fission products are retained within the fuel pellets and are responsible for the high levels of radiation and associated heat output of the spent fuel, which requires the fuel to be cooled initially in large water-filled basins or pools. The fission products are more radioactive than the minor actinides, and therefore decay much more rapidly. After several months the radioactivity of the spent fuel has decayed enough to allow the fuel to be more easily handled, either for reprocessing to extract and recycle the remaining fissile material, or for interim storage pending final disposal as waste. In the longer term, the heat output and the radioactivity of the spent fuel is largely due to the presence of minor actinides.
The moderator - how neutrons are slowed down
In a thermal reactor, neutrons emitted during fission must first be slowed down before they can be captured by fissile nuclei and provoke further fissions. This slowing down is done in the moderator, which must be a material made of a light element, but also one that does not readily absorb neutrons, which would remove them from the flux of free neutrons in the reactor. Ideally, the neutrons should just bounce off the nuclei of this light element and in the process lose some of their kinetic energy, but still remain available to provoke further fission reactions. Hydrogen, deuterium or carbon nuclei, in the form of ordinary water (H2O), heavy water (D2O), or graphite, are used as this moderator material. Nonetheless, neutrons may still need to collide with the nuclei of the moderator up to around 100 times before they are 'thermalised', or slowed down enough to be captured by a U-235 nucleus, thereby sustaining the chain reaction.
Control rods - how the chain reaction is tamed
To control the chain reaction, all reactor systems incorporate a material that absorbs neutrons i.e. removes them from the neutron flux in the reactor before they can provoke further fissions. Cadmium and boron are ideally suited to the task, as their nuclei are highly neutron absorbent (they are said to have a high neutron absorption 'cross-section'). Control rods made of these materials are inserted into the nuclear reactor to control the power output of the reactor or to shut down the chain reaction completely; withdrawing the rods from the core restarts the reaction or increases the power level. Often two sets of rods are used, one that is highly absorbent that shuts down the reactor immediately when inserted into the core, the other less absorbent that can be used to control the power level or for 'power-shaping' (equalising operating parameters across the core).
Current power reactors must be operated in a so-called critical state, whereby the neutrons produced through fission are exactly balanced by those being lost by whatever mechanism (provoking more fissions, absorbed by the fuel but without causing fission, absorbed by the structural material or moderator, or escaping from the reactor entirely). When critical, the reactor can be operated over a range of power outputs, effectively from zero power, representing very low neutron flux levels, up to the reactor's maximum power rating, equivalent to very high neutron flux levels. This is the most efficient operating mode of a civil power reactor, i.e. it is operating at its normal operating temperature and maximum power.
When U-235 undergoes fission, it releases neutrons very quickly; these neutrons are referred to as 'prompt neutrons'. If prompt neutrons only were released during fission, controlling the chain reaction would be very difficult, if not impossible. Fortunately, after a delay of a few seconds to a minute, the fission products themselves begin to decay, releasing additional neutrons called 'delayed neutrons'. There is only about 1 delayed neutron for every 100 prompt U-235 fission neutrons, but this is sufficient to make the control of the reactor much easier.
It is also possible to design a nuclear reactor that operates in a sub-critical mode, in which more neutrons are lost from the system than are created in fission, provided there is an external source of neutrons to compensate for the losses. Such an external source could be an accelerator that fires a beam of protons onto, for example, a lead target in the reactor, thereby creating a shower of so-called spallation neutrons. Such Accelerator Driven Systems (ADS) are considered to be particularly efficient at burning, through nuclear transmutation, nuclear waste as part of a strategy known as partitioning and transmutation (P&T).