This special interest group is where you can bring thoughts and ideas on renewable energy. 


You need to be a member of Energy Central to access some features and content. Please or register to continue.


Hydrogen Storage in a World of Wind & Solar

image credit: A Storengy Hydrogen Storage Concept

The article below is written by my colleague at the Green Power Academy, Grey Cells Energy & Speaker at the World Hydrogen Congress, Dr John Massey.

However often you’re told hydrogen is ‘the most common element in the universe’, this doesn’t change the fact that we don’t find it in elemental form. Hydrogen isn’t a primary energy source. We can’t drill or dig for it; instead we have to make it. So it becomes a store of energy that we’ve supplied elsewhere.

Renewable electricity can be one of these other supplies. Solar and wind in particular are becoming cheaper. In increasing numbers of countries these two sources are already the cheapest ways to generate electricity. Sometimes new such power projects even out-compete existing fossil-fuelled ones. Not surprisingly, most energy forecasters assume continued strong growth in the installed capacity of solar and wind.

Their downsides are well-publicised, boiling down to the fact that they supply energy when the weather allows, not necessarily when it’s most helpful (and valuable) for the overall power system. Building large capacities doesn’t guarantee electricity supply at times of peak demand, and it risks having too much at times when demand is low.

So pairing wind and solar with hydrogen seems like a match made in heaven.

Using these energy sources to produce hydrogen will ensure it is ‘green’. Having this green energy stored as hydrogen allows it to be used later, ‘on demand’. That demand could be to generate power again or even to meet a different energy supply need, such as transport or heating.

Power system flexibility

We already have energy storage in power systems and it is seeing rapid growth.

Most grid storage these days is pumped hydro. However hydro schemes require suitable sites, long development and construction times, and investors with both significant money and patience. So new projects are rare.

Instead the growth in electricity storage currently comes from that nemesis of many hydrogen fans: lithium-ion batteries.

Batteries are rapidly installed, at a variety of scales in all manner of locations; from giant grid batteries to domestic units. Their roundtrip (electricity-to-electricity) energy efficiency is high, at 85-90%, and other performance metrics are improving even while manufacturing costs are plummeting.

In South Australia, batteries have significantly lowered prices for very short-duration grid balancing (‘ancillary services’). In southwestern US states, solar-plus-storage plants are outcompeting gas-fired ‘peaker’ plants in competitive tenders.

Further ahead, ‘vehicle-to-grid’ (V2G) technology offers the prospect of aggregating batteries within large fleets of electric vehicles into sizeable grid-balancing resources. The same can be done with thousands of individual domestic or commercial-scale ‘behind-the-meter’ batteries. Examples of such aggregation already exist and many more are on the way.

Beyond lithium-ion, flow battery companies aim to overcome concerns around the former’s degradation, safety, longevity and materials supply. And beyond batteries, other storage technologies are regularly proposed: using heat (in various media) or cold (e.g. liquified air) or gravity (e.g. concrete blocks on cranes). Many will fail to cross the ‘valley of death’ to commercialisation and will disappear back to obscurity. But some will find market niches.

Using flexible demand to balance power systems, many large commercial end-users already adjust their electricity use: usually incentivised to lower it at times when supply is scarce or expensive. These tend to be end-users whose demands are driven by heating or cooling (in a well-insulated building, heaters or refrigeration units can be stopped for short periods without significantly impacting on temperatures inside). At smaller scales, ‘time-of-use’ tariffs are making their way into residential segments, encouraging the time-shifting of applications from water heating to electric car charging.

In short, it’s clear that the market for ‘power flexibility’ solutions, energy storage included, is a crowded and highly competitive one.

So where does hydrogen fit in?

Time & scale

The examples above have something in common: they provide energy storage or flexibility over quite short timescales.

Some batteries for grid ancillary services can discharge for only thirty minutes. Demand response programmes rarely expect end-users to forego demand for longer than an hour. Solar-plus-storage ‘peaker’ projects usually have three to four hour storage capacities. Though getting longer each time, the largest lithium battery projects currently under development will discharge for around five to six hours.

These are not theoretical limits, but limits set by the economics and practicalities of the solution. Even with plunging costs, the ‘per kWh’ costs of battery storage remain relatively high. There’s no point adding more hours of storage – at more cost - unless a service specifically requires it and the buyer of that service is prepared to pay for it.

Converting electricity to hydrogen and back has a poor round-trip energy efficiency, 35-40% at best; and easily less if its round-trip journey involves more steps. So we are wasting much more of that wind and solar energy we started with. How much economic loss this represents depends on the price of the energy when put into storage. Nevertheless, the lower roundtrip efficiency of hydrogen storage compared to battery means the former only makes sense where it has economic advantages that outweigh the energy losses.

To be competitive, hydrogen electricity storage will need to be much cheaper to build and operate, or must generate much more value from each unit of the energy it discharges (or a combination of both).

So where might hydrogen go, where batteries and other solutions fear to tread?

It isn’t difficult to dream up scenarios where converting renewable energy to hydrogen could be helpful.

Most focus falls on the economics of scale: storage over longer durations and much bigger volumes, where the modular stacking up of battery-filled shipping containers ceases to be practicable or cost-effective. It remains to be seen where this scale ‘tipping point’ will lie. However few believe batteries will make sense for storing the outputs of many 10s of GW of renewable power over multi-day time periods (to discharge during several days of low wind, for example). Or, bigger still, to balance seasonal differences in energy supply and demand.

Some even propose that hydrogen bulk shift renewable electricity geographically, on international scales: collecting oversupply in one region and sending it to another, where undersupply exists.

How can hydrogen be stored, in large quantities, for long periods?

The cost and energy intensity of liquifying hydrogen, plus evaporative losses in storage, mean this isn’t a great solution for large-scale, long-term hydrogen storage. Conversion to and from ammonia is similarly inefficient. There are a number of solutions for storing hydrogen within or converted to other chemicals. But most are not yet commercial, or are too problematic or inefficient to provide near-term, grand-scale options.

At large scales, we’ll be storing hydrogen as a compressed gas.

Salt caverns are already used to store natural gas (and have been also used to store hydrogen). They operate at pressures anywhere between around fifty and two-hundred bar: lower pressures mean lower compression costs, but require bigger storage volumes. In regions where suitable geology exists, construction costs and leakage rates are both relatively low. More information will be covered in the upcoming World Hydrogen Congress with speakers from  businesses such as Uniper and Storengy.

Other potential underground storage sites are depleted oil or gas fields. Gas fields offer lower capital costs compared to salt caverns and potentially very high capacities. Potential problems include leakage rates and issues with hydrogen-munching bacteria, but there is a lack of firm understanding until such storage is tested. Like salt caverns, this option depends on local availability.

Hydrogen can be stored anywhere in metal containers or pipelines, above or just below ground. Such solutions are hampered by larger investment costs for high volumes but can better ensure the purity of the stored hydrogen. Pipe storage has been used for natural gas storage since the 1980s and construction is relatively simple. For metallic containment though, the phenomenon known as hydrogen embrittlement requires different material choices and properties than is the case for natural gas, adding to costs.

For natural gas, transmission and distribution pipelines themselves act as large-volume stores for limited periods, by ‘line packing’ (increasing the pressure). Hydrogen looks likely to be blended with or ultimately replace natural gas in some networks, so this option could play a role too.

Who will pay?

Large-scale, long-term storage of clean energy as hydrogen sounds useful. Containment options exist. So why not just get on with it?

The answer is simple: because storage costs money. Someone has to pay to build, to operate and to finance the appropriate infrastructure.

In order to pay for this, the value of energy sold out of storage has to be higher than the value of that used to fill it. Round-trip energy inefficiency widens this required value difference, lower lifetime system cost reduces it.

Even if discharged energy is valuable, when sold on a ‘per-unit’ basis there must be enough throughput to provide a revenue stream sufficient to pay for the system (in particular its capital recovery). So the required sale price depends on how often the service is needed: rarely and it must demand a hefty premium.

Like batteries, hydrogen storage businesses are likely to need to familiarise themselves with ‘revenue-stacking’: using their costly infrastructure to compete for multiple revenue streams across various timescales and customers, while managing the contractual obligations and technical design considerations this will require. Simple ‘arbitrage’ business cases are already challenging for batteries and will be for hydrogen too, compounded by its lower round-trip efficiency.

In the case of seasonal storage, a business case may have to think about the challenge of highly seasonal cash flows, including risks to these (a warm, demand-reducing winter in the UK, for example). In practice, seasonal storage may be dependent on specific policy and market mechanisms, ones providing revenue assurance for those able to guarantee peak time capacity. The UK already runs capacity market auctions for this purpose; but low prices show how little value this service commands at present.

Since converting hydrogen to electricity is not efficient (at best 50-60%), one way to reduce the round-trip efficiency disadvantage of hydrogen (and the associated economic losses) is to not complete this round trip at all.

Instead, many believe a much more valuable approach is for hydrogen to act as an ‘energy vector’. Cheap, clean renewable electricity can be stored as hydrogen, then this hydrogen sold for direct use in market segments where other decarbonisation options may be challenging: heavy transport or industrial applications, perhaps heat.

Can we depend on ‘surplus’ renewable power?

For filling storage in the first place, it seems reasonable to assume that renewable electricity will be available very cheaply if oversupplied - as is the expectation more regularly, as larger capacities of cheap solar and wind power are built.

Nevertheless, it’s wise to test this assumption and recognise that hydrogen producers won’t be the only buyers of cheap electricity.

Demand response will eat up some of it: electric vehicle chargers, water heaters, heat pumps and other ‘smart’ devices will choose to take advantage of cheap power to minimise running costs for their owners. As ‘electrification’ of energy systems proceeds, the impacts on the shape and flexibility of demand curves could be profound.

Interconnectors, a growth area, will take some of our ‘surplus’: electricity will flow internationally between under- and over-supplied systems, driven by price signals.

Rather than give their product away cheaply, renewable power plants themselves will increasingly build storage into their projects: wind and solar power will become more ‘dispatchable’.

Even if only designed to address daily supply/demand timescales, each of these outcomes could eat up significant chunks of ‘oversupply’. It could be many years before these options ‘run out of road’.

If, when and where significant oversupply still remains, some output will simply be curtailed; perhaps even large amounts.

As renewable electricity costs less to generate, so too does the value loss through curtailing it. Curtailment may seem wasteful and unsatisfactory. But it can be cheaper to do this than to build, finance and operate the grid capacity or hydrogen production, storage and transport infrastructures required as alternatives. So curtailment can be a rational economic choice.

In summary, be cautious around the boldest assumptions you see around ‘surplus’ renewable electricity, in particular its regularity, scale, availability and price.

Not everyone is thinking big

Creating business cases and building infrastructures for large-scale hydrogen storage will take time. Perhaps, in the near-term, there are smaller-scale and less challenging market entry routes?

Some companies certainly think so.

In Vårgåda, Sweden, a 172-flat housing complex already lays claim to being the world’s first self-sufficient example of its type. It achieves this by generating all the energy needed for the year, for both power and heating, using rooftop solar PV. Hydrogen, produced from excess solar and stored in a pressure tank, is used to manage fluctuations in demand.

At an even smaller scale in the residential sector, GKN are piloting in Austria the use of a metal hydride storage solution to avoid the safety issues around storage at high pressures and to achieve higher energy densities (for more compact storage tanks, smaller than equivalent capacity batteries).

AFC’s hydrogen storage and fuel-cell power generation system aims to allow the deployment of fast charge-points for battery electric vehicles, while avoiding the expensive local grid upgrades needed in some places.

Such projects may prove to be very niche, or even uncommercial. In the short-term though, these smaller, more specific use cases at least present value propositions that are better constrained, more predictable and easier to test than goals of very large-scale, but far-off hydrogen storage.

That’s not to dismiss the latter or fail to recognise the large-scale supply and demand challenges that future renewably-powered electricity systems face. The long-term prognosis for hydrogen is positive. It’s simply to recognise that large-scale, long-duration hydrogen storage requires big investments in the face of very uncertain future demands and multiple competitors.

It needs brave, long-haul investors backed by clear policy and market supports in order to drive it forward.

Further Reading & References:


  1. A summary of UK winter power capacity, highlighting why growing more solar and wind doesn’t help, without other system balancing mechanisms, with the goal of ‘keeping the lights on’: (Oct 2019)
  2. This long document on hydrogen’s role in decarbonisation includes some numbers on why ‘surplus’ wind energy may be a small contributor to hydrogen supply in the UK (see page 76): (2018)
  3. A good technical review of hydrogen storage methods, including various emerging and non-commercial ones (warning: best suited for the more scientifically-minded): (May 2019)
  4. A nice presentation overviewing the issues around hydrogen energy storage, from given at the 8th Oxford Energy Day: (Oct 2019)
  5. Some examples of pilot projects in Europe looking at energy storage via hydrogen: (June 2019)
  6. An example of solar power and its impact on the electricity market (and prices), in this case the infamous ‘duck curve’ in California: (March 2019)
  7. Little renewable power curtailment these days occurs because of oversupply at the system level. Instead it is usually a local issue, caused by lack of grid capacity (this example data is from Germany). (March 2019):
  8. The example of a Swedish housing block using hydrogen paired with solar power: (Dec 2018)
  9. Residential scale hydrogen storage using metal powder technology? An example from Austria: (March 2019)
  10. The example of a hydrogen-powered EV charging station solution for locations with limited grid connection: (Dec 2019)
  11. Some companies are looking to scale-up emerging methods of hydrogen storage, in this example ‘LOHCs’ (liquid organic hydrogen carriers): (Aug 2019)


Nadim Chaudhry's picture

Thank Nadim for the Post!

Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.


Matt Chester's picture
Matt Chester on Feb 11, 2020 10:47 pm GMT

 Curtailment may seem wasteful and unsatisfactory. But it can be cheaper to do this than to build, finance and operate the grid capacity or hydrogen production, storage and transport infrastructures required as alternatives. So curtailment can be a rational economic choice.

This would be a moving target though, right? As renewable capacity grows so too would the total amount of potential curtailment and thus the economic incentive to do something more useful with it

Nadim Chaudhry's picture
Nadim Chaudhry on Feb 12, 2020 6:37 pm GMT

I think this does actually go far beyond curtailment. It's actually about achieving scale so the economics of pure play renewable projects set up exclusively to pump gas and not be grid connected works. This bascially should be cost competitive with natural gas with cheap electricity like $20pMWhr + a 50% reduction on electrolyser platforms from $700 pKW to $350 pKW (achievable by 2030?) + a higher carbon price of circa €50 per tonne. 

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »