Icarus’ Legacy — Episode 6: Energy Storage and Hydrogen

I believe that water will one day be employed as fuel, that hydrogen and oxygen which constitute it, used singly or together, will furnish an inexhaustible source of heat and light, of an intensity of which coal is not capable. Jules Verne, “The mysterious island”, 1875

(Google Slides version more suitable for large screens here)

In the previous episodes

The ingredients of our net zero recipe include a large dose of clean and renewable energies: the traditional hydropower and geothermal, and more and more solar and wind power, which will be the dominating force of this decade, given their broad applicability, and their reducing costs.

But alone, these sources cannot be a solution: their intermittent nature means that power grids will still need a solid base of reliable energy throughout the day, which doesn’t depend on the weather. If we want to gradually remove coal and natural gas power stations from this role, we will have to invest in nuclear power, and some of the “advanced nuclear” designs promise inherent safety, reduced costs and the resolution of a number of issues that have historically been associated with this energy.

Modern biomass can also play a role in the displacement of fossil fuels, by replacing them like-for-like in a number of cases in transportation, residential and industrial sectors through biogas, biofuels and solid biomass. But its expansion is debatable — we shouldn’t use more land or forests, even when managed sustainably, because they detract from biodiversity and food, and have the potential to create damage to other areas of society. The area of waste (municipal, agricultural and industrial) for bioenergy generation is very promising, and should be expanded.

In this episode we investigate two other tools at our disposal, which are in the early stages of their development, and are receiving a lot of attention: energy storage and hydrogen can support clean and renewable power generation by smoothing intermittent production to match demand, and by addressing a number of cases which cannot be resolved by using electricity, like aviation and other heavy transportation, or industrial production that requires high temperatures, such as cement, steel and some chemicals.

The full fifth episode is here, the fourth is here, the third here, the second here.

What is missing in the energy mix?

What is missing in the energy mix?

In the best possible scenario, the next 5–10 years will see a massive growth of electrification in all our society sectors, and significant replacement of fossil fuels with cleaner alternatives.

Electric vehicles will be more and more common in our cities, buildings will be heated with heat pumps, and many industrial processes will switch from fossil fuels to electric appliances. This increase in electricity use will be supported primarily by an increase in power generation from wind and solar energy, and by a solid base provided by nuclear plants. At the same time, solid biomass, biofuels and biogas will continue to provide a low-carbon alternative in those cases where combustion is still required (e.g. for shipping and aviation fuel or to convert fossil fuel power plants to modern biomass).

This should be doable in most countries, and hopefully in the largest energy consumers.

But unfortunately this wouldn’t take us very far, and the impetus of clean energy transformation would hit major roadblocks well before 2030.

No matter how much capacity we add in solar and wind, we will face intermittent, unreliable supply due to seasonal or weather patterns.

The likelihood of advanced nuclear to attract major new investments, become available and make an impact in most high-energy consumer countries is relatively low.

And sectors like cement, steel and chemical productions cannot switch to electricity, as it cannot generate the very high temperatures they need (not at a reasonable price, anyway).

So what is missing in the energy mix that can take us to the next level? Our hopes, and also massive investments, rest in energy storage solutions and hydrogen.

They are still quite expensive and in most cases far from clean solutions, but they will be dearly required, and in massive scale, if we want to go beyond the late 2020’s in our energy transformation efforts. The clock is ticking, and breakthroughs are required.

Let’s see where we are.

Energy Storage

The production of large quantities of renewable electricity from intermittent sources like solar or wind, or from other sources that can be subject to seasonality and weather, like hydroelectric, creates the requirement for large scale storage.

Storage can help match electricity demand with production, by allowing the accumulation of energy when production exceeds demand, and by releasing the stored energy to the electricity grid when demand outstrips production.

Even more importantly, when energy storage is incorporated in a renewable power plant, it makes this plant “dispatchable”: a precious (and hence better compensated) resource for the electricity grid operators, which can plan and fine tune supply in the best way and in real time.

Energy storage also gives a level of flexibility that makes a plant go beyond its nominal capacity when we hit high demand (and high prices). Even with nuclear energy, as we’ve seen for some advanced nuclear plants, we will make best use of energy storage (e.g. in molten salt) to reduce its output during low load, something that is normally not possible on a traditional nuclear plant, and discharge the energy, boosting generation, during high demand.

In other cases energy storage is provided by the grid itself, so during low demand times it can collect many intermittent sources (e.g. from rooftop solar panels, and non-dispatchable wind farms), and reduce energy waste by charging some (often far away) storage, for later use.

Storage: a Cost or an Investment?

It would be easy to tag Energy Storage as an extra cost attributable to the introduction of renewable sources, especially at a time when costs are still very high.

Some storage technologies are not quite ready to scale, still expensive or hard to apply in certain regions, with polluting production processes, and a lot of complexity potentially pushed towards the electricity grid. We are experimenting with many possible solutions, many of which are already in operation, but no one is emerging as a clear winner. We will probably need many of these solutions, depending on the context.

Grid complexity is a big deciding element: the more load balancing problems are resolved peripherally, close to energy production or energy usage, the less complex and expensive will need to be the long-distance transmission grid. So power plants, but also homes, businesses, communities, which are all transforming into “prosumers”, with their renewable production coupled with their increasing energy consumption, should better plan for some local storage.

So if we look at the overall system, we can see energy storage as an investment, as it:

  • Reduces waste of energy from curtailment, during high renewable production/low demand
  • Reduces the need of “peaker” fossil fuel stations, during low renewable production/high demand
  • Reduces or defers the need to scale transmission grids, by introducing local storage capabilities at power plant, community or residential level.
Source: IRENA

Electrical Batteries

The most obvious form of storage for electricity comes in the form of electrical batteries, which today use Lithium-ion technology in most cases. Over the past 10 years there have been massive improvements in battery technology, allowing for more energy density per weight, and faster recharge and discharge times. We are still in the early days of application of this technology at utility scale, and prices are quickly coming down.

At utility-scale, batteries typically offer 2–4 hours of coverage when discharging at full power, so they do offer a useful intraday compensation for solar plants or wind farms, but they are not yet a useful solution for cases where we need days of continuous output.

Tesla has been innovating in battery technology, both for its electric vehicles and for its utility storage offering. In 2017 Tesla installed in Australia what was the world’s largest lithium-ion battery storage facility, using its PowerPack modular technology, for a total capacity of 100 MW and storage of 129 MWh. Tesla claim they can scale to 250 MW / 1 GWh , capable of supporting entire cities or regions to compensate the peaks and valleys of renewable energy production.

Even larger scale projects are already live or in their implementation phase, like the Vistra California plant, which boasts 300 MW capacity and 1.2 GWh storage, and has been switched on in December 2020, to become the largest in the world. The company has already announced the construction of a new plant, twice this size. There’s other “feel good” aspects as these projects are housed in converted fossil fuel plant sites, which benefit from existing connections to the grid, improved the landscape with low profile buildings, and removed the pollution from the air. More and more, we will be able to see, feel and smell the difference that conversions from fossil fuels to renewable energy bring.

Cost of Electrical Batteries

There are two ways to look at the cost of utility-scale batteries:

  • Initial investment (purchase price): this is especially useful to give an idea of how prices have dropped in recent years
  • Levelized cost of electricity (LCOE): as for any other source of electricity, we can look at costs over the whole life cycle, as a better comparative metric. Larger batteries that offer higher power output will improve massively on this metric

In the period 2015–2018 there’s been a reduction of around 70% in purchase price, from over $2,000 per KWh to $625 per KWh and more recently, prices have dropped further to under $200 per KWh, and are predicted to go under $100 per KWh. The other aspect affecting LCOE is the durability of batteries, driven by the degradation from charging cycles. While technologies are improving, we know that the useful life time of a battery is a few thousands full recharges and discharges, typically corresponding to 10–15 years of operation, when operating them in full charges and discharges in the 10% — 90% range.

The LCOE has dropped to around $187 / MWh in 2019 and under $150 / MWh in 2020. At this price point, in several solar farm projects, the addition of suitable storage adds a premium of around $5–6 per MWh, which effectively allows it to be dispatchable, and demand a higher energy price from grid operators. This makes solar plants compete in flexibility with traditional fossil fuel or hydroelectric plants.

The large increase in production of lithium-ion batteries, especially for electric vehicles, is creating an interesting recycling opportunity: a car battery pack might become unsuitable after 5–10 years, if its capacity drops to under 80%, but it may well have many more years of useful life as energy store in a power plant, where capacity per weight is less of an issue.

Another aspect to consider for electric batteries is their use or rare earths, as we’ve also seen for solar panels and wind turbines (particularly for the inclusion of permanent magnets) and other metals, including large quantities of nickel, cobalt, graphite and lithium. The increased awareness means that supply chains are being improved, and recycling is starting to be on the agenda in most countries, including the improvement of regulations that force battery producers to take responsibility for the full life cycle of batteries, from production to disposal.

Pumped Storage

If electrical batteries are naturally coupled with solar or wind farms, pumped storage is a very efficient solution for hydroelectric power stations as it can be seen as an extension of a traditional plant, with an added lower level reservoir and pumps to push the water back uphill.

Hydropower is not intermittent, but it does depend on weather and seasonality, over relatively long periods of time (weeks/months) so a storage solution is not used to compensate for peaks and valleys of hydro production (which can already be tuned up/down very easily) but rather to compensate for wider grid requirements: pumped storage can draw energy from the grid when energy is very cheap and pump the water from the lower to the higher reservoir, and later generate energy when demand and prices are higher.

While energy losses of this cycle can be around 20%, the typical price differentials between high and low demand, and the important storage function provided, more than justify the economy of these plants.

The main advantage of pumped storage is its very large capacity: a single reservoir can store tens of GWh of energy, in the form of potential energy of its water vs. the lower reservoir, to be released when required, and capacity of multiple GW. The world combined storage capacity of pumped storage plants is over 9,000 GWh . It would cost almost $2tn to build electric batteries of the same storage capacity at today’s costs. It is no surprise that these plants are currently by far the largest energy stores in the world’s electrical systems.

The disadvantage of pumped storage is that not all countries and regions can afford to have such systems: large quantities of water and suitable high and low reservoirs with large drops in elevation but limited distance are required. For this reason they are naturally built in mountainous regions at latitudes that guarantee reasonable quantities of water flows.

The other disadvantage is that they often compensate production variability of plants that are far away from them, so they require large scale transmission grids. In general, well-designed electrical grids should benefit from locally balanced energy production, which can be predicted, programmed, dispatched and can serve nearby communities with as much autonomy as possible. In this way it will not overload connections across very distant points.

Hydrogen for Energy Storage

Hydrogen production through electricity and electrolysers can be seen as a form of energy storage, although a relatively inefficient one. It can be coupled with intermittent sources of energy like solar and wind, so that hydrogen is produced when electricity demand is low, and it can later be used to generate electricity via fuel cells, when demand is high.

If we compare this storage solution to electrical batteries:

The advantages of hydrogen are that:

  • We can store virtually unlimited energy — we just need sufficiently large containers for hydrogen — a very cheap technology
  • Hydrogen could be reused locally to generate electricity, or transported to other sites for different use, like in transport or to fuel industrial processes. Or a combination of the two

The disadvantage is that the cycle of production of hydrogen via electrolysis and the transformation of hydrogen back into electricity via fuel cells is very inefficient and can lose over 50–60% of the input energy, respect to around 10–15% of loss in charging and discharging electrical batteries

With hydrogen, we don’t need to invest in electrical batteries, but electrolysers are also a significant investment. The cost of both technologies is reducing quickly, but as of today the cost per KW are fairly compatible (the rate of power at which we can produce hydrogen, or at which we charge a battery).

Molten Salt Storage

The Concentrated Solar Power (CSP) technology, where solar energy is used to produce heat, often leverages Molten Salt as energy storage, by heating it to very high temperatures, and using the heat, potentially after hours or days, to generate electricity. Molten salt is heated until it reaches a liquid state and stored away in isolated containers, which can later pump their content back to generate electricity as in normal thermal plants via hot steam and turbines. With proper insulation, tanks can be maintained at usable temperatures for over 1 week.

The same technology can be coupled with other types of electricity production, and generate molten salt at relatively low temperatures, so it doesn’t necessarily require a CSP.

Important cases include advanced nuclear reactors, which by their very nature produce heat, but all renewable energy production, where production cannot be fine tuned, could be in principle coupled with heat generation and a molten salt reservoir.

The advantage of this technology is that the amount of storage available can scale quite significantly, and can be used over long periods of time.

Getting Creative with Energy Storage

Many experimental technologies are being tested in real cases around the world, and they could offer creative but still very effective solutions without posing geographical constraints, and typically with clean production processes:

  • Hot Basalt Rock: stones are ground to a fine powder, so they can be heated via hot air at hundreds of degrees and constrained in isolated containers, for later use to generate steam and electricity. Siemens Gamesa have been innovating in this sector.
  • Liquid Air: air is pumped at high pressure to liquid state, either in massive underground chambers or in special containers, which can later release their pressure to operate turbines and generate electricity. What is interesting is that spent oil or gas wells offer suitable chambers for this technology, given their massive size and increasing availability. This technology doesn’t have special geographical constraints and the first large scale plants are being built.
  • Lifts: very large weights can be lifted to heights with different technologies, from cranes that build a massive tower from stone blocks, to train carriages that are moved up the hill on old railways. Gravity does the rest of the work: when the potential energy is released and the weights come back ground level, they activate generators that provide energy back to the grid

Energy Vault is a Swiss-based, global energy storage company specializing in gravity and kinetic energy based, long-duration energy storage products. Energy Vault’s primary product uses a multi-headed crane to store energy by stacking heavy blocks made of composite material into a tower, capturing potential energy in the elevation gain of the blocks. When demand for electricity is high, the crane lowers these blocks to the ground, with the motors functioning as generators and delivering electricity to the grid.

What is interesting of all these solutions, from heated salt and rocks, liquid air and lifts, is that they can scale, they can be used in every geography (unlike pumped hydro) they are environmentally-friendly (unlike batteries), reasonably priced, and have limited degradation.

Electrical batteries have the advantage that they waste less energy in their recharge cycles, but the relatively dirty production materials and high degradation mean that they might face a stiff competition at utility scale, once all these technologies mature to their fullest.

Hydrogen, while less effective as energy storage (due to high energy waste and costly electrolyzers and fuel cells) provides other opportunities, as we will see in the next section.


Hydrogen is one of the most abundant elements on Earth, but it very rarely occurs in free state, so it’s not considered a primary source, and it must be extracted and produced from other compounds.

There are five ways to produce hydrogen, which have been given colors, to give an idea of their different “merits”.

Black and brown hydrogen is obtained from bituminous coal and lignite respectively, via a gasification process, which releases CO2 and carbon monoxide. Grey hydrogen is produced from natural gas, via a steam reforming process, which also releases CO2. Blue hydrogen uses the same process as the Grey one, but the CO2 emissions are captured and not released to the atmosphere. Finally, Green hydrogen is obtained from water, using low-carbon electricity with a process called electrolysis.


Grey hydrogen is what is produced today in most cases. It’s relatively cheap, but very bad due to its emissions. Blue hydrogen is a viable option, but it’s very expensive due to the carbon capture element. The ideal scenario is when Green hydrogen is produced with cheap electricity from renewable sources, so we don’t generate CO2 in any part of its cycle.

Hydrogen today is mostly used in industrial processes to produce ammonia or fertilizers, or to refine petroleum.

But for decades it has been considered as a candidate for the decarbonization of several sectors, from heavy industrial processes, to transportation and residential use.

Hydrogen Hopes

Why is Hydrogen important for the energy transformation? For a number of reasons: it can

  • replace fossil fuels in industrial processes where we need high temperature (e.g. in the production of cement or steel) as its combustion doesn’t generate GHG emissions
  • replace fossil fuels in heavy transportation cases, like airplanes and ships, and perhaps lorries and buses
  • be produced using cheap electricity from renewable sources
  • be a convenient way to store energy, as once produced, it can be easily stored and transported using normal gas infrastructure, or even re-transformed into electricity through fuel-cells (reverse electrolysis)

The growing interest has recently come from the availability of cheaper and cleaner electricity from renewables, and the cost reduction of electrolysers. The investments that are converging on hydrogen production at scale are promising, especially when coupled with solar and wind production

Advert for the Toyota Mirai hydrogen fuel-cell vehicle

Other cases have been debated for long time, and while there’s some scope for short term implementations, hydrogen is less likely to become a long term solution:

  • Fuelling light vehicles: battery operated electric vehicles offer a better solution than fuel cell vehicles, due to high energy loss through electrolysis and fuel cells, but also because of recent improvements in battery technology: the argument of longer range is fading away
  • For residential heating and cooking, as a direct replacement of natural gas: electrical appliances like heat pumps and induction hobs offer a better long term solution. In the short term hydrogen could be blended, in small quantities, with natural gas
  • To extend the lifetime of natural gas power plants, by burning a blend of natural gas and hydrogen through modified turbines. Long term, as it takes a lot of electricity to produce hydrogen, it won’t make economical sense to use hydrogen to produce electricity

One of the noteworthy initiatives is from the European Union, as part of its green deal, a 2x40GW production capacity is planned for 2030 . By realizing this electrolyser plants to produce green hydrogen, about 82 million ton CO2 emissions per year could be avoided in the EU (about 2% of current emissions — a large amount for a single initiative). The total investments in electrolyser capacity will be 25–30 billion Euro, creating 140,000–170,000 jobs in manufacturing and maintenance of 2x40 GW electrolysers. The idea is that, utilizing the renewable energy potential of Europe, Northern African countries and Ukraine, and the existing natural gas networks, hydrogen could be produced very cheaply, and transported throughout Europe to feed a large distribution network for transport and industrial use.

Hydrogen for the Transport Sector

In transportation, aviation and shipping, but also large vehicles such as lorries, buses and trains (replacing diesel stock), could benefit from the combustion of hydrogen or from fuel cell technology. These vehicles wouldn’t necessarily require a widespread network of hydrogen distribution, which instead would be a must if we wanted to address light vehicles.

There are still some technological issues to overcome for large scale aviation, and aircraft projects tend to be very long. Considerations about volume and weight are very important: hydrogen has energy density per weight 3 times larger than conventional kerosene (so we would reduce fuel weight) but energy density per volume, even when hydrogen is in liquid form, is 4 times lower than kerosene. So hydrogen would take much more space and require high pressure tanks, which would make it unsuitable for storage in wings (where kerosene tanks are normally fitted) and would make the fuselage abnormally longer or larger.

Also, very importantly, it is still not conceivable to have electric engines provide the necessary power and thrust to lift large aeroplanes, so the idea of hydrogen for flights requires transforming internal combustion jet engines, so they can burn hydrogen, instead of kerosene. Fuel cells and electric motors can only be considered for light, non-commercial aircraft and prototypes have already been proven.

Airbus has presented a zero-emission hydrogen concept for a large aircraft that could enter operation in 2035.

On the other hand, ships, buses, lorries and trains have less of a problem with space, and we already have viable solutions in all cases. Electric batteries might soon compete in the market of lorries and buses, but fast refuelling and long range could remain an advantage for hydrogen.

Hydrogen buses have been in use for several years all over the world, hydrogen trains have been in operation in Germany since 2018, and deployment in other countries is planned for 2021. Hydrogen lorries made by Hyundai have been in operation from 2020. In the naval industry, real examples of hydrogen-powered yachts, carrier ships and even submarines have been demonstrated and deployed over the past 20 years.

The only ingredient that we have been missing to see the massive growth in these sectors is the availability of cheap green hydrogen, but that’s where the investments are coming. And the demand will be massive.

So far, car and light vehicles manufacturers seem to have preferred battery electric vehicles to hydrogen fuel cell technology, for the reasons explained below, but this might change as technologies evolve, and availability of hydrogen improves.

The advantage of hydrogen fuel cells vs. petrol internal combustion engines is the absence of toxic and polluting emissions and the much higher energy efficiency (2–3 times better) measured on weight, although it does require high pressure and larger volume tanks.

The advantage vs. fully electric vehicles is in the lower time of refuelling/recharging (5 minutes vs. hours, in some cases), but otherwise electric vehicles enjoy a much higher energy efficiency, which translates in much lower operating costs.

The process of producing hydrogen from electricity via electrolysis, and then generating electricity via fuel cells to feed an electric motor, will lose as much as 65–75% of the original energy. In comparison, an electric car will only lose 10–15% of the energy taken from the grid to charge and discharge the battery. Electricity is also widely available in both residential and commercial settings, in city roads and around highway networks, so electric vehicles will only need incremental improvements to the existing grid, to scale electricity delivery and deal with peak demands. A roadside network for hydrogen would instead require very large investments.

A hydrogen fuelled vehicle would also normally be equipped with an electric battery, to allow for improved car performance (torque/power) but this battery would be much smaller than the ones used in fully electric cars, and act as a buffer between the fuel cells and the electric motor.

Hydrogen for the Industry

A sector in which the introduction of hydrogen could prove essential for our net zero objectives, is the industry. There are sectors of the industry including steel, cement, other building materials and some chemical productions which require very high temperatures, which today can only be achieved by burning fossil fuels. In principle hydrogen can offer the same capability, and the issue is only its production cost and production at scale, which are both predicted to be solved in this decade.

To appreciate the scale of the problem, cement production alone is responsible for 8% of global GHG emissions. This is due primarily to the high temperatures required in kilns to produce clinker, the key ingredient of cement. While many alternatives are being considered for more environmentally friendly cement, including:

  • Using less clinker in the cement mix
  • Carbon capture and storage, including capturing it into the cement itself
  • Using less cement in buildings (!)

Hydrogen could be a relatively easier way to replace fossil fuels in the production, and several feasibility studies have already been conducted. This UK study modelled a transition to different fuel mixes including biomass and hydrogen, and while it showed technical feasibility in relatively short time for a net-zero production of cement, it highlights a number of challenges to be overcome, and the very high costs implied by the transition.

The Power production industry itself could already benefit from the use of hydrogen: a natural gas power plant can be fitted with a new generation of turbines that operate with mixed natural gas / hydrogen fuel. Recent versions can support a 50–50% mix, massively reducing emissions.

Hydrogen for Residential and Commercial use

Another area where fossil fuels are used in large quantities, is the heating of buildings and water through natural gas. While moving to electrical appliances including heat pumps, induction hobs and electrical heaters is seen as the best long term approach, due to its efficiency and compatibility with renewable sources like wind and solar, it is also true that we currently use a very large gas network, which will probably remain in use for a long time.

In this respect, experiments are underway to use hydrogen on the same network, initially blending 10%-20% of hydrogen with natural gas, which shouldn’t require any upgrade to the gas network or to the appliances already in use, like gas heaters and boilers. But it is unclear what could be the long term effect of this, including the impact on pipelines (embrittlement of metals and possible damage to other materials used for gas pipes).

There is no doubt that sectors like industry and heavy transportation, where electricity is not an option, could benefit from hydrogen — this is the most promising hope to effectively displacing fossil fuels.

In sectors where electricity is instead a viable option (e.g. light transport vehicles and residential and commercial buildings) the case of hydrogen as a transition fuel is much weaker. It could reduce use of natural gas for heating in the short-medium term only if there were no other side effects (compatibility with existing gas networks and gas appliances).

Key to all this will be the cheap production in large scale of hydrogen using renewable electricity sources, and its distribution to end users, which should be able to leverage the investments made over the past decades for natural gas.

Recent predictions see the cost of green hydrogen production (through electrolysis) to drop significantly over the next years, to be much cheaper than the cost of hydrogen production from fossil fuels (grey or blue hydrogen, in particular).

Renewable sources of energy and a good dose of nuclear power can provide the foundations for the energy transformation, but other ingredients will be required to allow for increased electrification, and to resolve problems where electricity cannot be a solution.

Energy storage at utility scale is a requirement whenever we introduce large quantities of renewable energy from intermittent sources.

Hydrogen can provide a solution in those cases where electricity cannot be applied.

The next 5–10 years should see both areas expand massively, with a range of new and enhanced technologies that, together with dropping prices, should see larger and larger adoption in all sectors of society.

No matter how fast we will progress in our roadmap to net zero, for many years to come we will continue to emit more GHG into the atmosphere than our planet can naturally recycle. But we have a very limited budget to do so, if we want to avoid the worst of climate change effects.

So while we reduce new emissions, we need to look at how we can also remove existing CO2 from the atmosphere. This is technically called Carbon Capture and Sequestration, and it will be the subject of the next episode.

In the next episode: Carbon Capture and Sequestration



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Andrea Ceccolini

Software developer, passionate about science, maths, sports and people