1.10: Thermal storage

One thing we will need a lot more of in the future is energy storage. Solar and wind don’t produce electricity all the time and also have seasonal fluctuations in power production. If we want to provide electricity 24/7 then we’re going to need to be able to store excess energy to use later.

One option is batteries but these often can’t get to the energy densities needed to provide energy to things like heavy industry. An emerging solution is thermal energy storage (TES). The basic idea is that you can turn the electricity into heat and store it in some medium, say some super high temperature liquid, then, when you need electricity you can use a heat engine to effectively reverse the process.

One overarching reason this approach could be useful is economic, or as this article in nature notes:

Current technologies, such as pumped hydroelectricity, are geographically limited and lithium-ion batteries (~US$80–100 kWh–1 capital cost) are too expensive for the multi-day storage targets (~US$3–30 kWh–1) needed to fully decarbonize the grid….However, the key advantage for TES is its potential for low cost (< US$20 kWh–1) at the gigawatt scale.

The reason TES could get so cheap is that almost all of the atoms in your storage medium are actively being used to store energy vs only a small proportion of atoms in something like a battery. Also, you are far less constrained in what materials you have to use than in something like a battery. From the same article:

Since the specific heat of virtually all materials is the same on a molar basis, at high temperatures, TES can make use of extremely abundant and low-cost materials that are impure or even recycled

Molton salt

One of the first types of TES was molten salt storage. Salts often have the advantageous property of having a low melting point and a high boiling point. This means that you waste less energy in phase transitions during heating and also don’t have to waste heat preventing your medium from freezing. HITEC salt used in solar plants has a liquid temperature range of 149 - 538°C. You can then take this molten salt and use it to boil water to turn a turbine when you need to convert it back into useful electricity. Molten salt storage has not massively taken off so far despite being one of the most established methods. One main reason for this is costs. The example in this course roughly estimates that the end result of storage costs currently comes in far at over $100, above the useful ranges we outlined at the beginning of this section.

Carbon blocks

New, lower cost, solutions to this problem are being developed. Antora is a recent company that aims to use blocks of carbon as its storage medium. Why carbon? They outline a number of reasons which broadly break down into, super cheap input, very high temperature range it can operate under (as a solid), these both contribute to a very simple and very cheap potential system.

On the input side carbon blocks (or prebaked consumable carbon anodes) are already created in huge volumes, and low cost, in the metals industry e.g for aluminum smelting. Using carbon blocks for TES therefore starts with a much cheaper storage medium than something like molten salt (in fact it is roughly 5-10x cheaper). As we can see in the image below:

specific heat capacity of different materials

We can get the energy that something can store using E = m*Cp*ΔT where m = mass flow, Cp = specific heat capacity and ΔT temperature difference. The great thing for carbon as we can see is that it has a very high specific heat capacity and it can also take much larger temperature changes than most other materials. (also note that if you are only going for low temperatures e.g up to 500℃ then molten salt has the advantage but this becomes overwhelmed by carbon’s abilities at higher temperatures.)

As mentioned above though what we care about ultimately is the cost per kWh of energy which we can get using Cost of storage medium / (Cp*ΔT). Given the low cost of carbon and the high energy it can store, Antora estimates it could get the price for energy storage down to $1/kWh which would put it over 50x cheaper than lithium ion batteries.

The carbon blocks can operate at over 2000℃ which creates another benefit that TES approaches like molten salt don’t have. At this temperature most of the heat energy actually comes in the form of radiation not convection or conduction. This basically means that for your heat recovery step you are dealing with very energetic light, which allows the use of a much more efficient heat engine.

Heat engines

Antora mentions that they have “developed a world-record-breaking solid-state heat engine that converts radiant heat into electricity with only a few micrometers of material and no moving parts”. While I don’t know the specifics of this system, recent research has come out on a very similar sounding setup which is very cool.
picture of heat engine
The set up is that you have something that is akin to a solar PV cell which captures high-energy photons. Remember because of the incredibly high temperatures we are using most of the heat energy comes out as radiation i.e photons.

While still experimental demonstrations have this new heat engine reaching efficiencies of up to 40%, which would put it above that of conventional steam turbines at ≈ 35%. The solid state nature is also crucial if you want to use the TES for very high temperature industrial processes as traditional turbines can’t operate above 2000℃.

It should be noted that the approaches above don’t solve things like seasonal changes in energy production as they have a useful lifespan on the order of hours not weeks. For durations of that length we may well need to turn to other solutions like hydro-pumping, which is already being done at some scale, or things like electrofuels which can remain untouched for long periods of time.