Heat Storage, a piece of cake?
Energy storage is a question that receives much attention. It seems that the storage of electric energy is the most important, judging by articles about batteries, both home batteries and large scale installations. Problems are the cost and availability of Lithium and other materials used in batteries, and the near impossibility to extinguish Lithium batteries which catch fire[1].
The emphasis on electric energy is surprising, because the main part of domestic energy use, in the Netherlands, is for heating. Therefore, it makes sense to store energy in such a way that it can be retrieved as heat. Is this easier and less dangerous than storing it in electric batteries?
The time-problem
At present, the main green energy generating device in (or: on) Dutch homes is the solar panel. It is attractive because it can make use of the electricity grid, and because electricity is a high and often uncertain cost in homes. The investment pays for itself in roughly six years.
The solar energy is generated when the sun is up, which it is for 8 hrs in winter and 16 hrs in summer[2]. Electricity is needed throughout the day, but mostly when people are home, i.e. not at work. The highest demand is outside office hours, including the evening, night and early morning. At these times,there is no or insufficient solar energy available. During the day, when production is at its peak, and people are at work or at school, demand is low.
Storage of electricity
The time-problem has given rise to home batteries (see this installment of Reden Makes Sense). Home batteries store electricity and deliver At present prices, the home batteries are not very attractive. Due to internal resistance, there is a loss of roughly 5% of charged energy (depending on charging rate)[3]. For this type of storage, a cheaper method, which requires less Lithium and Cobalt, would be very welcome.
However, is it sensible to store electricity? Is electricity the main energy need? It might be easier to store energy for future heat requirements.
Heat versus electricity
In 2022, the average household, according to the CBS[4],used 2640 kWh of electricity, produced 230 kWh of electricity, and used 980 m3of gas (equivalent to roughly 9800 kWh). For simplicity, we ignore the fact that a part of the electricity is used by heat pumps to provide heat, and assume instead that the division of energy between electricity and gas (2640 vs. 9800) represents the division between electricity use for non-heating purposes and heating. This means that heating our homes requires 79% of the domestic energy consumption.
If heat could be stored cheaply, without time limits and efficiently, it would make sense to convert the excess electricity generated by solar cells during the sunny summer days to heat for use in the cold, dark winter months. This is exactly what the Salt Battery promises. It is being developed by Cellcius[5].
Salt Battery
The salt battery does not store heat simply by heating a material, which then is kept in a well insulated container for future use (and hoping the heat is needed before it has leaked away. Instead, it makes use of an interesting property of salt, which allows storage of energy without changing the temperature.
Salt is hygroscopic; leave it in a damp environment and it will absorb water. This happens spontaneously, which should mean that the absorption of water releases energy. The reverse process costs energy, so drying the salt costs energy.
Common salt(NaCl) is not the best salt for the job, as the amount of energy per kg stored when it is dried is small. Researchers at the Technical University of Eindhoven think they have found the best candidate, also a cheap, abundantly available and simple compound: Potassium Carbonate (K2CO3) . To give an idea of how safe it is, K2CO3 can be an ingredient of baking powder[6]. By the way, this is a dangerous argument, because danger depends on dose. The safety data sheet[7] of K2CO3 warns, that:
“It is a severe irritant of the eyes, skin, nose and throat. Ingestion of large amounts is corrosive, and may result in circulatory collapse and death.”
The safety of the system, therefore, depends on the low risk that it ends up in any part of the body, not on the wholesome nature of the material.
How much volume is taken up by the salt battery? In the video, Joey Aarts mentions that 20 modules with 20 kg of salt each can last an average household the whole winter. We can take this to be in the order of 5000 kWh (total gas use equivalent to 9800 kWh, minus total use for hot water and cooking, minus some heating in spring and autumn), which gives an energy density of roughly 250 kWh per unit of 20 kg, and 12.5 kWh per kg of salt, or (again roughly[8]) 21 kWh per litre. In contrast, this paper gives a value of 0.27 kWh/l. This is a factor 78 less! [9]Does this mean the average household needs, not 20, but 1575 units of 20 kg of salt? That would mean a battery of 7.5 x 2.5 x 1 m, and bring the material cost (according to the same paper) to more than €30,000 (at 2017 prices)! For comparison, the energy density of domestic heating oil is 10 kWh per litre, and Li-ion batteries up to 0.7 kWh/l.
A competing technology is thermal heat storage, for example Cesar. It claims a storage density of 250 kWh per m3, which is virtually the same as the value given in the scientific paper cited above, of 0.27 kWh/l for the salt battery. Cesar gives the heat density of Lithium batteries as 150 kWh per m3, which is much lower than the range of 250 – 693 according to Wikipedia[3]. Cesar does not take into account the higher value of electricity (see ‘Multiplication trick’ below).
If we put the energy density in ascending order (ignoring Cesar) and take the density of heating oil as 100 %, then the list becomes
- Salt 2.7%
- Li-ion 7%
- Heating oil 100%
This means that the volume is 37 times bigger than an oil tank containing the same amount of energy.
By the way, the regular readers of Reden Makes Sense will wonder how these energy densities compare to that of iron powder[10]: Iron powder would rank above heating oil, with 140% (1 kg 7.39 MJ, 1/0.8/8.5 litre, -> 14 kWh/litre).
Multiplication trick
Converting electricity to heat can be done with a very high efficiency. A simple resistive coil has an efficiency of very nearly 100%. However, there is a better way; electricity is a high value form of energy (in thermodynamic jargon: it has a high exergy content), whereas low temperature heat is a low value form of energy (only a small part of it could be recovered as electricity). Using a heat pump, the amount of heat which can be stored for later use can be 4 or 5 times higher that the excess electric energy. For this, the salt battery has to be big enough, of course.
The same trick can be used with the Li-ion battery, but at the other end. The electricity from the battery can be used to drive a heat pump, which multiplies the energy by a factor of 4 or 5. If a Li-ion battery is used for heating, it can therefore have an effective heat storage of 25 – 30% per volume of the heat contained in heating oil, or roughly ten times as much as the salt battery[11].
Transport
A great advantage of the salt battery is that it can store low temperature heat, and can be transported. This makes it possible to re-use waste heat for domestic heating.
Conclusion
The salt battery sounds like a very good idea. The energy density is less than would be concluded from the 'average family' claim in the video, so it cannot store sufficient heat for the whole winter. Nevertheless, it may still be attractive as a buffer. Its merits are doubtless better than the “cars that capture carbon dioxide[12]."
Numbers are important, because they make the difference between useful and useless, between possible and impossible, between attractive and unworthwhile ideas. At first glance, the salt battery seems a piece of cake, but a look at the numbers reveal that some claims made about it should be taken with an unhealthily large pinch of salt.
--------------------------------------------------------------------------
Thumbnail: tue
[1] The present remedy is to place complete EVs in a water filled tank to cool them until the batteries have burnt out.
[2] The values are 16 hrs and 44 minutes on 21 June, and 8 hrs and 43 minutes on 21 December(2024); the value varies constantly over the year and depends on location: Hours of daylight
[4] Energieverbruik particuliere woningen;woningtype en regio's
[5] Presumably, this is a play on 'Celsius', the unit of temperature named after Anders Celsius. I understand the extra 'l', but why the second 'c'?
[6] Baking powder
[7] Data sheet of baking powder
[8] Depends on bulk density and voidage of the powder. The bulk density is 2430 kg/m3 , so if the voidage is 0.3 then the density of the powder is 2430*(1-0.3) = 1700 kg/m3
[9] The Cellcius System Architect, Pim Donkers, clarified the situation by explaining that the unit mentioned in the video was intended to work as a short term buffer, storing energy for assisting the heat pump, depending on heat demand,and the availability/price of (solar) electricity. He expects a density of 0.16kWh/litre in the near future.
[10] Reden Makes Sense: Is iron the new hydrogen
[11] For thermodynamic fanatics: heat too can be used to power a heat pump, but only if the heat is released at a high temperature. This seems to rule out the salt battery, but not the iron powder. This seems an interesting track for research,but for now, we do not quantify it.
[12] Reden Makes Sense: Cars that capture carbon dioxide: fact or fairy tale?
