100% Renewables: Is Massive-Scale Energy Storage for Renewable Energy Possible?

25 minute read

Updated on Wed Mar 03 2021

Renewables are likely to become very popular

By now you are probably sick of reading ‘the sun doesn’t always shine and the wind doesn’t always blow’. We are too. But, sadly, that’s a real problem.

Storing energy is a difficult task. It’s so difficult, that the electric grid is designed such that electricity is consumed the instant it is produced. Think of it like this: when someone turns on a lamp, a power plant somewhere increases its output slightly.

Energy storage - why and how?

To get a rough idea, we can categorize energy storage into three timescales:

  • Level 1 - Seconds-minutes: Helps when a cloud temporarily shadows a solar farm or when electricity demand peaks unexpectedly. You might see terms like ‘load balancing’ to describe this.
  • Level 2: - Hours-days: This form of energy storage allows us to generate electricity when it’s sunny or windy and use it whenever we want to, for example at night. Energy storage at this level is often called ‘bulk storage’.
  • Level 3: - Weeks-months: Energy storage at this level works with seasons so that maximum electricity is generated during the summer, and stored for the winter. This is important because in winter the sun doesn’t shine as much (unless you are near the equator).

There is a clear pattern in what attributes each type of storage system should have:

Storage needs

Level 1 storage (easy)

California electricity demand May 7th 2020

The circle shows the type of (often unexpected) ups and downs that the level 1 storage has to deal with. Level 1 storage can already be done with various types of batteries.

Today's batteries are fantastic sprinters, but we need a marathon runner for level 2 and 3 storage. In particular, we require:

  • High scalability (availability of materials and space)
  • Low cost at scale
  • The ability to hold charge for long enough

Level 2 storage (this is hard)

Why we need Storage across hours

Solar panels produce most electricity during the day, but household demand peaks in the evening. Level 2 storage would help us store electricity produced at noon (when the sun is out) for use in the evening.

With some baseload zero-carbon electricity from nuclear, we wouldn’t need as much storage because the lowest points of the curve would be nudged up.

Moreover, by using controllable electricity sources like hydropower and natural gas (a fossil fuel) during peak electricity demand, we might get to 90% clean electricity plus 10% gas, with relatively small amounts of storage needs.

Why baseload and dispatchable power help

Level 3 storage (harder)

The appropriate amount of seasonal storage depends on where you are on Earth. In general, the farther you are from the equator, the larger the difference in incoming sunshine between the summer and winter months. This affects how much solar energy we can harvest.

Need for seasonal storage depends on where you are

Solving level 2 and level 3?

Let’s look at some of the alternatives to batteries and hydrogen. (Remember that charging means putting energy in and discharging means getting energy out.)

  • Pumped Hydro: Charge by pumping water up a hill. Store it in a reservoir. Discharge by letting it flow down to spin a turbine.
  • Flow Batteries: Like normal batteries, but the cathode and anode are liquids.
  • Thermal Storage: Big refrigerator. Charge by using electricity to heat one substance and cool another. Discharge by using the temperature difference to spin a turbine.
  • Liquid Air: Charge by cooling and compressing air until it becomes a liquid. Store that in a tank. Discharge by letting it turn back into a gas and spin a turbine.
  • Compressed Air: Charge by compressing air. Store it underground. Discharge by decompressing and letting it spin a turbine.
  • Flywheels: Charge by spinning a big rotor really fast. Discharge by converting rotational energy to heat or electricity.

We will briefly discuss how each of these works before reaching a conclusion.

1: Pumped hydro

Pumped hydro

97% of electricity storage today is through pumped hydro. In well-suited locations, pumped hydro is relatively cheap and can facilitate long-term storage.

Pumped storage is different from hydropower, as hydropower works by having a river fill a reservoir - here we just store energy and don’t need a river.

But like hydropower, pumped hydro has its issues. Many countries don’t have enough mountains to build pumped hydro storage (remember that pumped hydro requires water to be pumped up a hill). Where we can build it, large areas of land need to be flooded and there is evidence that this causes significant methane (CH₄) emissions.

Still, pumped hydro should be used where environmentally acceptable. In the end, we have to weigh these downsides against the alternative: burning coal and gas.

2: Flow batteries - liquid battery tanks

Flow battery

Flow batteries use liquid anodes and cathodes. Although they are made from different materials than Lithium-Ion batteries (normal batteries), the principle is similar:

  • Charge by pushing electrons from the cathode to the anode.
  • Discharge by letting the electrons flow back through a circuit.

Flow batteries present some advantages over normal batteries:

  • Storage capacity (tank size) and power (electrode surface area) are independent from each other. Need more storage? Make the tank bigger. Need more power? Increase the size of the area where anode and cathode meet.
  • Potentially longer lifetime.

3: Thermal storage

We can use electricity to heat things up with nearly 100% efficiency. This heat can be stored for a few hours and then converted back to electricity using a turbine.

Heat-pump for thermal storage with molten salt and chilled liquid

A potentially more efficient way to do thermal storage is what your refrigerator does. It uses electricity to cool one thing and heat another. The device that does this - in your fridge and in the proposed storage technology - is called a heat pump.

Here, the cold is stored in a liquid and the warmth is stored in molten salt. You might remember ‘molten salt’ from the chapter on nuclear energy!

The ‘reverse’ of a heat pump is a heat engine. Heat engines can turn a temperature difference into electricity. This ‘temperature difference’ is simply the temperature of our molten salt minus the temperature of the cold liquid.

4: Compressed Air - cheap and environmentally friendly?

Compressed Air Energy Storage

In this form of energy storage, air is stored underground at a high pressure and can later be released to spin a turbine.

This method requires large underground reservoirs that don’t leak.. Not every hole in the ground can do this. Theoretically, there are many suitable regions, but it's not clear if the technology can scale as far as level 2 or 3 would require.

Only two plants have been built so far, even though people have been working on this since 1940.

5: Liquid air

Liquid Air storage

To liquidize air, it needs to be compressed and cooled. The liquid air is kept in a storage tank. To get the energy back out of the system, some of the liquid air is evaporated.

Gaseous air fills more space than liquid air. When we let the liquid evaporate, we can use the expanding air movement to spin a turbine that generates electricity.

6: Flywheels - spin, spin, spin

Flywheel energy storage

We can store energy by spinning a rotor (called a ‘flywheel’) at very high speeds in a vacuum.

What form of energy is this?

The rotor speed is increased or decreased to store or deliver electricity. The rotor’s speed can be adjusted with almost no delay, which makes them useful for level 1 storage. However, flywheels cost far more than other storage technologies.

It's all about cost

Let’s look at numbers for a developed nation like the US. The US uses 3.95 PWh of electricity a year. That’s 12 MWh per person per year. Dividing by 365, we get 32 kWh per person per day. The cost of electricity is $0.18/kWh in the US.

The equation we’ll use is:

Equation for cost of electricity with storage

The total cost factors in the loss of energy from storage. Conversion of energy from one form to another usually involves loss of energy to the environment, so we need to produce (and pay for) more electricity to start with.

Prices vary based on region and provider. The $150/kWh cost we used before is the raw production cost. With the company’s profit, on-site installation, assembly, shipping, and R&D costs included, the same company (Tesla) that produces batteries for $150 recently sold a huge battery pack to a wind farm operator for $511/kWh.

The average expected price is a little lower at $350/kWh. Working with this, we get

Battery levelized cost (per year)

Let’s do this for all of the technologies we've discussed. (We'll compare two Lithium-Ion battery types: NCA and LTO. Similarly, we'll look at two flow battery types.)

Evaluating all the energy storage technologies

So, are any of these cheap enough?

Experts think that $20/kWh for scalable and location-independent storage with a very high retention rate would be cheap enough for it to be widely deployed.

That’s why we need innovation and government support. Flow batteries are expected to cost 66% less by 2030 while getting longer lifetimes and higher energy retention rates.

We are trying to solve an unsolved problem. Picking only one technology today would be unwise - that’s why research funding is going to many companies in parallel.

Either way, we will likely combine multiple solutions to provide storage in regions with differing geologies, climates, and economies.

One thing is clear: For anything near 100% renewable energy, we need cheap, large-scale storage solutions.

Next Chapter