Energy can be stored in the binding-energy of the crystallization of water (or other substances). In the case of water in particular, this is about as much energy as would be needed to heat the water by 80 °C within one phase (i.e. without a phase transition), for example from 10 °C to 90 °C.

Since the storage of ice requires nothing more than good thermal insulation, it can be used to store energy quite efficiently.


To make the energy usable again later, a thermal engine is rather inefficient because the thermodynamic efficiency is too small due to the small temperature difference between the ice and the ambient temperature. An optimistically calculated outside temperature of 30 °C in summer (or realistically in warm continents) would result in an efficiency of just η=1-T cold/T warm=1-273K/303K ≈ 10%.  This is not worthwhile.


BUT as a basis for air conditioning (for example in hotels), ice storage could become quite efficient. Excess energy could be introduced into "ice storage" at times when it is not being used, and then withdrawn from ice storage at times when the air conditioning is running. Cooling with ice is a really simple technology.


Thermodynamicists also design and build ice heaters that can extract heat from the stored water in winter and thus heat a building, freezing the water from which the heat is extracted. In summer, the frozen ice can then be used to provide cooling.

Very well done by Viessmann:

Further explanations can be found on the Internet:

This sounds surprising at first, but in a way it reminds us of so called absorber refrigerators, which cools, being provided with energy by heat.


Due to the smaller temperature differences, we are therefore only talking about heating and cooling systems at this point, but not about the possibility of re-converting the heat energy into mechanical and/or electrical energy, because the latter variant would be too inefficient due to the second law of thermodynamics (see above, efficiency ≤ 10%).