Does one Ultimate Energy Storage Solution Exist?
… … … 25 minute read … … …
If you spend any time googling the topic of ‘decarbonising the energy sector’, it won’t take long before you come across an article highlighting energy storage technology as being the key to adopting renewable energy. Here I label it as ‘The Holy Grail’ of the renewables revolution – but I must admit, this isn’t an entirely original title…
At least the unoriginal nature of the ‘Holy Grail’ phrase demonstrates that this really is a critical subject, which is why I’m covering it as one of my first blogs. I will be discussing some methods we have to store energy and how there is unlikely to be just one ultimate solution but rather an intelligent diversity of storage facilities.
In the last blog I described how the energy supply is controlled in order to meet demand and how it’s a bit like controlling a sensitive tap. To extend this analogy further – you can imagine multiple different taps representing different energy sources. Some taps are easy to adjust the flow of and others are a bit stiff and take some time to adjust. Imagine you’ve got all these different taps around a bath tub full of water; the plug has been pulled out so the water is draining away. It’s your job to keep the water level at a constant height by adjusting all of these taps accordingly. Furthermore, (bear with me) the draining hole, which represents demand, is constantly changing in size, so sometimes the water is draining faster than other times. Nuclear power would represent a tap that’s almost completely stiff, with water flowing out at a constant rate. Coal power represents a fairly stiff tap so it takes a lot of time to adjust the flow rate. Gas is a brilliant tap that can be easily adjusted. Unfortunately, the renewables tap is completely broken, spurting out random jets of water at different rates, and this clearly makes your job keeping the water level constant a lot harder. Energy storage would basically fix the renewables tap and make the rate of flow easily controllable. If this mended renewable tap was also made sufficiently powerful, you could even turn off the gas and coal taps permanently – which incidentally, have been providing dirty water this whole time.
So how do things stand with energy storage today in 2020? Well, it’s rather insignificant. This year in the UK, pumped storage (which at present is far and away the dominant energy storage technology), made up a tiny 0.6% of the energy mix. Renewables have made up on average 26% of our energy mix in the past year. So we know that we can have at least a quarter of our annual power coming from renewables with practically no storage. But as I mentioned in the last blog, we’re starting to see the limitations of increasing renewables in the UK with no storage. There is a growing need to turn off some of our renewable generating potential when the favourable weather conditions mean we can be generating more electricity than we can consume. Turning off wind turbines etc. is expensive, so to avoid this we’re also seeing negative electricity prices, incentivising industries to use up some of this excess electricity, because that’s cheaper than turning things off and on. If we aim to keep increasing the renewable energy component of our energy mix, this wastefulness will become increasingly problematic and hinder their profitability. So these facts clearly illustrate that it is high time to invest more in energy storage.
Bottom line is, renewables are commercially attractive and so they are quickly going to dominate many countries energy mixes in the coming years and will therefore need cheap and effective energy storage technologies to accommodate them – that much is inevitable. So let’s have a look at what these technologies are going to be. I listed some in my last blog and I’ll quickly list them again here, just as an aperitif… pumped storage, battery storage, compressed air energy storage, flywheels, chemical energy storage, high temperature thermal energy storage, electromagnetic storage, amongst several others. If I tried to describe them all now, this blog would unfortunately be far too long, so I’m just going to pick a few that are particularly promising and illustrate the varying storage requirements.
Starting off with pumped storage. This can’t not be discussed because it is overwhelmingly the dominant storage technique used currently, accounting for 95% of global electricity storage in mid-2016. It works by pumping water from a lower reservoir to a higher reservoir by using the excess energy during periods of low demand or high renewable supply. When demand is high, and/or supply is low, this stored gravitational potential energy can be utilised by sending the water back down to the lower reservoir through a turbine, to turn a generator and supply the necessary additional electricity. So it basically balances out the variable renewable supply, which is the aim of all energy storage systems. (If you want to know the proper jargon, the act of storing electricity when demand is low and supplying power when demand is high, is known as load balancing.)
In the UK, pumped storage has accounted for 0.6% of the energy mix last year, or alternatively 0.17 GW. To put that value into some kind of context, the average UK electricity demand is about 30 GW and is at its highest about 50 GW. So if this 0.17 GW doesn’t sound like much, that’s because it isn’t really. Currently this pumped storage is used, not so much for storing excess renewable power, but more for being able to meet sensitive spikes in demand. In the previous blog I reference a long established small surge in power when EastEnders finishes and everyone boils the kettle for a cup of tea (yes even electrical transmission networks can have stereotypes). Pumped storage has the ability to release its energy potential very quickly and rapidly increase the total electricity supply. This sensitivity is highly valued by grid operators, allowing them to match the sharp spikes in demand at short timescales and this is the primary benefit of pumped storage in the UK at present.
If we want to use pumped storage to satisfy the requirement of supporting renewable intermittency, we will need to implement it at a larger scale. How large a scale exactly? That’s a very difficult question to answer. Energy storage is only one component of the overall energy flexibility required to support renewables. There are also ideas of smart grid technologies, improved interconnections and the diversity of renewables (see last blog). But I had a google and the best I could find was an estimate of what the total thermal/electrical storage we would need (in the UK) to meet an 80% reduction in carbon emissions, calculated using something called the ‘Storage and Flexibility Model’, (the details of which are provided in this link). This value should be treated with caution but they suggest that 1400 GWh (or 1.4 TWh) of storage capacity is required. So if we now compare this with the largest capacity pumped storage facility in the UK, ‘Dinorwig Power Station’, this has a total storage capacity of 9.1 GWh. Assuming these estimates are correct, we’d need over 150 of these facilities to meet the UK’s total energy storage needs. Which seems worryingly high to me.
But this pumped storage solution is clearly geography dependent. For countries such as Norway, with great hydropower potential, and other mountainous regions – this technology has great value. A recently published paper, by Hunt et al., 2020, estimates the global potential of pumped energy storage capacity (below a cost of 50 US$ MWh−1) to be 17.3 PWh, or the equivalent of about two million Dinorwig Power Stations. This value is only an absolute upper limit of what’s implementable, and the world certainly doesn’t need 17.3 PWh of storage capacity any time soon. The more pragmatic use of this study was to show how hydrological potential is “unevenly distributed” and varies across the world, with some countries having great potential and others having very limited potential. So ultimately pumped storage will be cheap for some countries and too expensive for others. This is one key factor which means that a single ‘ultimate’ storage method isn’t realistic and that there’s a need to innovate other solutions.
Pumped storage is basically just gravitational potential storage, simply using water as the elevated mass. Before moving on, I just want to very quickly mention how it’s not the only gravitational storage method. This short Guardian article discusses an exciting energy storage project in development which utilises Britain’s old disused mines for an energy storage technology. It works simply by hoisting a load of 12,000 ton weights up and down these vertical mining shafts as a cheap form of gravitational energy storage. Wrench them up when there’s excess renewable energy – drop them down and spin a generator when there’s a deficit. No need to drill massive 800 metre shafts or build 800 metre high towers, most of the hard work has already been done. This is a perfect example to illustrate the benefit of creative solutions which adapt to what potential is at our disposal. Such methods may be limited in scale but when combined with many other innovative, cost effective, small scale storage technologies, they can provide an overall substantial and cheap energy storage potential.
Hydrogen – the most abundant element in the universe. A fact that is often used to introduce the topic of hydrogen storage. An accurate but fairly useless fact as we can’t harvest the whole universe for its resources. But hydrogen is also abundant on Earth – in the form of water, or H2O. So to acquire the pure hydrogen we simply need to extract the ‘H2’ from the oxygen (the ‘O’). This is done by a very simple process and one which you probably did at school. That is, the electrolysis of water – involving not much more than two pencils, a bucket of salty water and an electrical supply. You can connect the graphite of the two pencils to a circuit, creating a pair of electrodes – a positive anode and a negative cathode. When you stick these charged rods into the water, electrolysis begins to take place. The H2O compounds in the electrically conductive salty water gain enough energy to separate into their component parts – the positive hydrogen molecules being attracted to the negative cathode and the negative oxygen molecules towards the positive anode. The hydrogen can bubble its way up through the surface and be collected. Thus, you have made hydrogen. Easy.
Hydrogen is also highly combustible and can therefore be used as a fuel to generate electricity. Once you burn the hydrogen to produce energy it simply combines back with oxygen in the atmosphere to produce the one and only by-product of water. If you used a renewable source of energy to create the electrode electricity supply, the hydrogen fuel can be considered as a ‘renewable fuel’.
So at this point you’re probably already one step ahead of the game. Yes, you can indeed use excess renewable energy to create hydrogen, which can be stored and burnt to produce more electricity when the renewables are not generating what’s required. This is a form of ‘chemical energy storage’ and is something that’s already done. It’s not exactly as I describe above, for example they don’t use pencils when making hydrogen through electrolysis on an industrial scale, but it’s still the same fundamental process. It’s important to be aware, 95% of hydrogen production at present is not made through electrolysis but rather a thermal process involving the ‘steam reforming’ of natural gas, producing carbon monoxide – and so this does not count as a renewable fuel! (Carbon monoxide is not directly a greenhouse gas but it can increase the abundance of greenhouse gases such as carbon dioxide and methane.) Furthermore, the electrolysis process used to create hydrogen at present is predominantly powered through non-renewable energy sources (which is rather frustrating). So be cautious that at present, a lot of hydrogen fuel really isn’t all that clean!
Anyway, returning to the benefits of hydrogen fuel made through renewable electrolytic processes as it should be. One great advantage of this energy storage technique over pumped storage is that you don’t need a mountainous region with great hydroelectric potential, all you need is the abundant resource of water. It also takes up significantly less space than pumped storage, with no need to create reservoirs, therefore avoiding as much heavy engineering and so avoiding the accompanying ecological impacts. It also shares one of the key strengths of pumped storage which is its storage timescale. Once the hydrogen has been produced it can be stored and will have a certain amount of energy generation potential. If the hydrogen is stored appropriately, this potential won’t reduce over time. The same thing can’t be said for batteries, for example – as we’ll see.
One place where hydrogen energy storage is coming into fruition is in the Orkney Islands in Scotland. The Orkney Islands have some of the highest renewable potential of anywhere on Earth, with lots of wind, wavy seas, and strong tides. They produce more renewable energy than their 22,000 population consume and so are a net exporter of electricity. They are currently attempting to increase their hydrogen energy storage in order to store this abundance of renewable energy and implement the renewable fuel in their infrastructure (ferries, boilers etc.) I recommend this YouTube video made by the great ‘Fully Charged’ channel, if you want to hear more about renewables and hydrogen storage on the Orkney Islands.
Hydrogen storage may never be used in countries such as Norway. Having so much hydrological potential, Norway generates 96% of their electricity from hydropower and so their energy mix is already effectively one large battery. But with hydrogen storage not being as geographically restricted as pumped storage, it will prove to be an important, cost-effective method for less mountainous countries. Predictions made by the International Energy Agency suggest Hydrogen fuel, created using wind power, will be cheaper than natural gas by 2030. So that’s very exciting!
Hydrogen can also be stored in a variety of ways making it a versatile solution for many of our low carbon energy needs, from transport to domestic heating to grid applications. Hydrogen fuel has the highest energy per unit mass of any fuel, but a very low energy per unit volume. If you hear the phrase ‘energy density’ – that usually refers to energy per unit volume; energy per unit mass should be referred to a ‘specific energy’). Hydrogen storage methods therefore focus on fuel compression, storing as much hydrogen in as small a volume as possible. One of the most ‘everyday’ considerations of energy density is in transport applications, as it ultimately determines the range of the vehicle. Hydrogen fuel cell cars currently have similar ranges to pure battery electric vehicles but they have limited refuelling infrastructure and are yet to become cost competitive. Whilst hydrogen vehicles seem to still have a little bit of catching up to do, hydrogen fuel used in home heating systems shows strong potential. Decarbonising electricity with renewables is proving to be increasingly successful, but renewable heating solutions (with the majority of heating in the UK currently powered by gas boilers) are proving harder to come by. This makes solutions such as hydrogen fuel cells exceedingly valuable for a complete decarbonisation.
For grid ‘load balancing’ applications, the hydrogen can be stored at a much larger scale. Salt caverns are a promising storage solution as they have historically been used to store oil and natural gas and so would also be suitable for storing large quantities of hydrogen gas under high pressure. According to one study, the storage capacity of one large salt cavern is between 65 and 210 GWh and in the UK there are roughly thirty large caverns. So this represents a massive potential energy storage capacity, exceeding the estimated 1400 GWh required by the UK. The same research group estimated a colossal 84.8 PWh of potential storage across Europe alone.
I think, given all of these facts, it’s very likely that hydrogen storage will be one of the premium renewable energy storage methods in the future, we just need to make sure it is in fact made from renewable energy sources! To point out yet another Holy Grail reference, one paper I found was entitled ‘Hydrogen Energy Storage: The Holy Grail for renewable grid integration’.
Battery technology has come a long long way over the past decade, largely owing to the associated constant market pull of laptops, smart phones and most importantly, electric vehicles (which all use lithium-ion battery chemistry). The graph below shows how the cost per kWh of lithium-ion battery storage has reduced over the past ten years, with an average 20% reduction in cost each year.
Extrapolating into the future, this cost reduction trend is expected to continue, falling to $100 per kWh by 2024 and by 2030 prices could be as low as $61 per kWh – an incredible 95% cost reduction from 2010. The cost of batteries is directly related to the cost of electric vehicles (EVs). The $100 per kWh mark is recognised to be an important threshold at which point price-parity between EVs and internal combustion engine (ICE) vehicles will be achieved, inevitably leading to a greatly increased market. The National Grid now expect there to be over 10 million EVs on the road by 2030 and 36 million by 2040.
You may be thinking, that’s great for the transport sector but what do EVs have to do with decarbonising the energy sector? Well unlike pumped and hydrogen storage, battery technology has great potential to contribute to ‘load balancing’ at a domestic scale. There are currently 38 million cars on the road today. Imagine that these were all electric vehicles, each with a battery storage capacity of 75 kWh. Multiply that by 38 million and the UK privately owned car fleet could have a total storage capacity of 2850 GWh – over double the estimated 1400 GWh of required storage. So any ideas of utilising this future storage potential would clearly be of great value. One concept in its early phases of implementation is that of ‘vehicle to grid’ charging. The idea being that all vehicles can charge from the grid at night when electricity demand is low and therefore when electricity prices are cheap and these car owners can sell this electricity back to the grid in the evenings when demand is high. And this discharging can happen on millisecond timescales which is a level of sensitivity greatly valued by the National Grid. So this is a situation where everyone’s a winner – saving consumer’s money whilst effectively balancing the grid. This idea is particularly attractive to big companies with large transport fleets, such as delivery companies. Their transport fleets could collectively act as a significant energy store, selling large amounts of electricity back to the grid and could even act as an additional source of revenue. As the grid becomes smarter, the UK’s collective transport fleet will inevitably act as an important ‘load balancer’.
In addition to EV batteries, there is an increasing uptake in domestic batteries such as the Tesla Powerwall 2, which is what’s shown in the above image. The idea of this product is fundamentally similar to the ‘vehicle to grid charging’ idea; it can be charged by the grid when prices are low and can then be used at periods of high demand, allowing you to avoid the most expensive electricity prices. But the particular target market of the Powerwall are customers who have an installed solar array. This is proving to be a profitable combination and allows consumers to exist largely off-grid, also making them invulnerable to power outages (not really a problem in the UK). They can generate their own power to be stored by their battery and used when electricity prices are high, if they ever require a top up, they can charge up with off-peak electricity from the grid. If during long, sunny summer days they are producing more power than they can consume, they can sell this power to the grid, effectively making them a mini power station. This is one form of what is known as ‘microgeneration’. It is something that is becoming increasingly common and will necessitate an increasingly sophisticated grid to manage. The idea of every house owning a domestic battery storage facility isn’t ridiculously farfetched however and is something that some people in the field advocate, including the company ‘Solo Energy’. They consider such a decentralised set-up, (which they call a ‘Virtual Power Plant’) as a practical and achievable solution for future energy flexibility.
Batteries are not only being considered as a domestic energy storage solution but also in centralised, grid scale facilities. Many of these systems, in existence and in development, are also of a lithium-ion composition. With the ever decreasing prices, this composition appears to be the most obvious choice. For such stationary energy storage facilities however, the mass and volume constraints are a lot less significant than for transport, so other, less energy dense battery compositions are also possible. One example is sodium-ion batteries – less energy dense than the lithium-ion make up but holds the potential to be cheaper, simply because sodium is a more abundant and accessible element than lithium. Material and mining limitations for batteries, as well as for other renewable technologies, are significant and is certainly a big topic worth returning to. To quickly highlight a couple of the key factors, firstly, the obvious economic consideration that technologies which only require abundant, easily accessible materials have greater financial potential. The second issue is a humanitarian one, with the extraction of certain materials such as cobalt, being intrinsically entangled with corruption, child labour and broader poverty issues.
Another drawback with batteries is their storage timescale. Whilst they can store and discharge power to satisfy the daily fluctuations in energy demand, the seasonal variations, (with energy demand being higher in winter) requires a longer timescale energy storage system such as pumped or hydrogen storage. Whilst these seasonal variations are not as significant as the daily ones, this is an acknowledged disadvantage of battery storage and is another reason why there’s not one ultimate energy storage system.
When looking towards the future of battery technology, the extrapolated cost reductions of lithium-ion batteries are exciting enough – but there also exists the potential for some major leaps forward. One revolutionary battery technology could be solid state batteries. (‘Solid’ because they use a solid glass electrolyte as opposed to the liquid electrolyte in conventional batteries such as lithium-ion.) They have the potential to be at least twice as long lasting as lithium-ion with higher energy densities, whilst being safer (a lower formation of dendrites) as well as suffering less from degradation with a greater total of charge/discharge cycles. Furthermore, they can be made from more abundant materials such as sodium and could have less expensive manufacturing processes, meaning they have the potential to become far cheaper than current batteries. It may all sound too good to be true, but this technology has received a lot of support from a man named John B. Goodenough –a legendary veteran in the field and one of the 2019 Nobel Prize winners for lithium-ion battery development, so not a bad person to be accrediting the technology.
…To indulge in a hypothetical for a moment, I think if I was given one billion pounds to spend on a single technology to help solve climate change, I’d probably invest it all in solid state battery technology.
Okay time to wrap things up, but I feel I should say, despite this being a very long blog, there’s a lot I have not discussed. I’ve not even mentioned thermal energy storage! But think I am pushing my word count a bit far (4000 and counting) and I believe these three main examples do a pretty good job of providing an overview of what our energy storage needs are.
So a few of the key takeaways. Firstly, some storage solutions are geographically restricted and some storage solutions are very unique to a certain environment. Secondly, the timescale of stored energy losses vary between each of the storage methods, (and energy demand varies on multiple timescales). Finally, storage methods can work at both grid scale and at a domestic level. These are three reasons which mean it is unlikely that we’re going to see one single, ultimate energy storage facility. Rather, there will be many storage technologies working in tandem, coordinated in a sophisticated way by the future ‘Smart(er) Grid’. With so many of these storage technologies showing rapid development, improved cost effectiveness and larger potential capacities, to me it seems worthwhile having continued investment in all of them. And I also hope that you were as encouraged as I was with all this promising development. It certainly seems like the energy storage requirements of renewables is not an awkward stepping stone to lay down and walk across on the path to a zero-carbon energy mix.
To conclude by returning to the big fundamental question of this blog site, that is – how do we stop burning fossil fuels? I occasionally hear ideas that we’ll only stop burning fossil fuels when there are no fossil fuels left to burn. This is simply not true. We’ll stop burning them when renewables become cheaper than them. This is slowly beginning to happen already, but if you were to point to one ‘thing’, one ‘Holy Grail’, that will really make this revolutionary transition happen… it is energy storage. To quote (believe it or not) the former oil minister of Saudi Arabia:
“The Stone Age came to an end, not because we had a lack of stones, and the oil age will come to an end not because we have a lack of oil.”
Sheikh Ahmed Zaki Yamani (2000)