There is often a clear difference between short-term and long-term storage needs. For instance, batteries in lawnmowers or electric vehicles have to be recharged every day, while batteries based on the same technology in watches and calculators can last for years before having to be recharged or replaced.
Looking at the past
The potter’s wheel is a classic example of a rotating plate storing kinetic energy. The plate stabilises the rotational speed and helps perfect the shape of the pots produced. Flywheels use inertia in a similar way, and when attached to irregularly rotating shafts – like those driven by explosion-type engines – can smooth out the shaft’s power output.
The storage time associated with such use of flywheels is a fraction of a second. And the use of flywheels has also been suggested for more substantial energy storage i.e. placing large flywheels horizontally underground can smooth out variations in renewable energy input from power sources such as wind.
The newer flywheel designs for applications like those above offer different strategies to balance the need for high energy-storage capacity, against minimal stresses in the materials used (i.e. with implications for the maximum rotational speed that can be sustained).
|"What flywheels do for mechanical energy, capacitors do for electrical energy - and probably at a lower cost."
And what flywheels do for mechanical energy, capacitors do for electrical energy - and probably at a lower cost. The electric current is used to build up assemblies of charge that can either be used to create delays in a circuit, or - by switching the current on and off - to store energy that can later be regained directly in the form of electrical energy.
Back to basics - energy forms
In principle, any energy store works with three energy forms: energy flowing into the device; energy stored in the device; and energy delivered out of the device.
Because the all-important round-trip efficiency must deal with conversion losses each time one energy form is converted to another, it is advantageous to keep the number of energy forms involved as low as possible, and not to lower the quality of the energy flowing through the storage device.
Lowering quality is actually unavoidable, if a high-quality energy form such as mechanical or electrical energy is converted into heat and later has to be re-converted into a high-quality energy form to satisfy the final user (as defined in the second law of thermodynamics.) The lower the temperature associated with the heat phase, the higher the loss. Flywheel technology is appealing because the energy is mechanical both for input, storage energy form and output, as are capacitors, where all the energy forms involved are electromagnetic.
Furthermore, storing electromagnetic energy could take advantage of superconducting materials where Ohmic losses are avoided. For example, a magnet could be charged by a superconducting coil with near-zero heat losses. But because known superconductors do their 'work' at very low temperatures, the system has to be cooled, and likely at a substantial energy cost.
Demonstration facilities already built, for example by the US Department of Defense and by the CERN scientific centre in Switzerland, allow slow charging but extremely fast releasing of the energy stored - essential in some applications. These installations are costly and there is a substantial economy-of-scale, which makes superconducting storage a large- and medium-term option.
Such installations are envisaged as having an important role in intercontinental power exchange through superconducting cables. The reasoning is that if the cryogenic technologies are already used in transmission, one could add value by using cryogenics for superconducting energy storage - aimed at load-management, or handling intermittent renewable resources such as solar electricity from desert locations (i.e. those of North Africa, as seen from a European perspective).
For batteries, the intermediate storage form is chemical, not electrical. This causes losses, and problems of electrode and electrolyte degradation.
There are two broad classes of batteries used or considered: conventional batteries with the chemicals stored inside the battery, and flow batteries - also called reversible fuel cells - in which the chemical substances are stored outside the unit.
At present, fuel cells of the alkaline type dominate the electrolysis market, while the reverse process - hydrogen to electricity - is only used in vehicles demonstrating a possible future use of another type of fuel cell (PEM) for road vehicles. Any kind of fuel cell or battery is characterised by the choice of electrode material, and the choice of electrolyte.
|"Storing electromagnetic energy could take advantage of superconducting materials where Ohmic losses are avoided."
Development has moved away from liquid towards solid electrolytes, and though the lifetime problems are similar for battery and fuel cell technologies, it seems that batteries are performing better than fuel cells at the moment. This is because neither PEM nor high-temperature fuel cells have yet reached their promised efficiency and durability, in contrast to advanced battery types such as lithium-ion batteries, which first out-competed lead-acid and nickel-metal technologies in the small consumer appliances sector (watches, mobile phones etc.), and later in products such as laptop computers, lawnmowers and hedge saws. And they now seem to be attaining the price and performance level required for hybrid vehicles.
This means that electrochemical storage technologies that used to be considered small-scale and short-term are now being touted as a solution to storage requirements on nearly any scale, including power utility applications - for example in the case of intermittent wind inputs to the utility system.
Any problems? Price of course, and round-trip efficiency still lurks around 75%. The difficulty is that losses are determined by electrochemical potential jumps across the storage device that cannot always be optimised in both directions, for storing and for retrieving power.
The examples above have been for storing and retrieving high-quality energy, and it should perhaps be mentioned in passing that for non-renewable energy sources such as fuels, those in solid or fluid form can be stored directly, whereas the gaseous ones (natural gas and hydrogen produced from renewable sources) need to be compressed. This leads to questions about the associated energy expenditure, not to mention the possibility of leaking gas.
The systems requirements make this storage option more interesting for large-scale stores. For example, low-quality storage - particularly low-temperature heat used in households and commerce - is typically under 100°C and sometimes substantially under this level. A heat store could be a hot-water tank of a few cubic metres, serving a household for a winter day (or two to three summer days) before it needs reheating; or it could be a communal pebble bed store, assisting a building-integrated solar heating system.
When moving from heat capacity stores to latent heat stores, an advanced system might use the energy absorption and release associated with phase change in a suitable material - such as Glauber salt (Na2SO4 with 10 crystal water molecules per salt molecule). And finally, it is possible to move from physical change to chemical reactions, selecting some that involve large amounts of energy added or released, depending on the direction of the reaction. A number of these systems have been tested in various configurations, from integration into the heating system of a detached house, to that of an apartment building or cluster of houses.
Because of the cost of solar collectors, the current economics in areas with a space heating demand favours smaller systems aimed only at supplementary energy during a (possibly extended) summer period. This case, as well as that of pure hot-water supply in areas nearer to the Equator, has made storage in water tanks the preferred technology.
The problem with solar heating is that solar radiation has a seasonal variation opposing that of heating demand, and the efficiency of solar thermal collectors also depends on the inlet temperature of the water circuit through the collector. This makes it difficult to select an optimal size for the heat store, because a small system has a low efficiency (high inlet temperature) during summer where the solar resource is largest, and a large store has too low an outlet temperature during the winter to be able to provide useful heat. One might add a heat pump to remedy this, but that is another capital-intensive component in a system that is already quite expensive.
About the author:
Bent Sørensen is a Professor at the Institute for Environmental, Social and Spacial Change, Roskilde University, Denmark.
He has written several books on the subjct of energy storage including:
- Renewable Energy (3rd ed., Elsevier 2004);
- Hydrogen & Fuel Cells (Elsevier 2005);
- Renewable Energy Conversion, Transmission and Storage (Elsevier 2007).