"We already have the technologies available today"

To date, the topic of storage has played a rather minor role in the energy transition in Germany. From when on, will this have to and will it change?
It made sense for us to focus on expanding renewable energies first. Unlike with renewables, however, we should not build as much storage as possible, but rather what is currently required. Every storage facility incurs additional costs, and its use causes efficiency losses.  In various studies, we have been able to show quite well that, in fact, up to a share of about 50 percent renewable energies, there is no real need for storage at all from the point of view of the power grid. And long-term storage is only needed beyond a share of 80 percent renewables.

Now, however, we are exactly at the point where storage facilities are necessary in order to get away from fossil power plants. Now, there are increasingly periods when conventional power plants are not in fact needed. However, in order to actually be able to shut them down during these phases, we need storage facilities for the instantaneous, primary and secondary reserve. This is necessary to balance out fluctuations and ensure grid stability. 

Another important point for me is that we always talk about storage. Viewed from the perspective of the power grid, however, we don't necessarily need storage facilities, but flexibility. In other words, if there is a lot of renewable energy, we need something to utilize the surplus power for a useful purpose. Furthermore, if there is too little electricity, we have to release power. This can also mean switching off loads. From a technical point of view, a storage facility is the ideal element for this, however, the task of balancing supply and demand can also be achieved in other ways.

In the political discussion, people also like to refer to future innovations that are necessary. Do the storage technologies for the energy transition still have to be invented?
No, the technologies are already available to us today. We have solutions for short-term storage with lithium-ion batteries and also with classic lead batteries. Long-term storage in the electrical sector is gas, i.e. hydrogen or methane produced from green electricity. Between these technologies we don't need that much more. When it comes to innovations, the question can only be: can we make the technologies more affordable in terms of lifecycle costs? Storage systems are still expensive, of course. Battery storage systems have significant investment costs that have to be amortized over their life cycle. Research must focus on finding alternatives for limited raw materials on the one hand, and on making the technologies more cost-effective in terms of lifecycle costs on the other. Investment costs, efficiency and service life, but also recycling, play an important role here.

Then let´s get specific: which storage technologies will be used for future supply security as well as grid stability?
Anything that needs to be balanced on a 24-hour basis is the domain of battery technology. Charging and discharging takes place every day, and that is where efficiency plays a major role. Therefore, you would not choose gas for a daily cycle, as the conversion from electricity to gas and back to electricity has an efficiency of only about 40 percent in the case of hydrogen as a gas. That means for every kilowatt-hour that will be returned, there will be a loss of one and a half kilowatt-hours. In comparison, a lithium-ion battery including converter achieves an efficiency of 85 to 90 percent, which is quite a different story.

On the other hand, we need a buffer for the so-called dark and wind scarce periods of up to three weeks. The energy system must, therefore, be able to supply itself from storage facilities for up to three weeks when fossil fuels are no longer available. From what we can see today, this task can only be performed by gas. In this case, the relatively low efficiency no longer plays such a decisive role. Instead, it is almost exclusively the investment costs that matter.

This can be calculated quite easily: if we use a very inexpensive stationary battery storage system, the installation costs amount to around 100 euros per kilowatt hour. If we assume a 20-year service life and need the storage unit on average once a year, then each kilowatt hour sold would have to earn an average of five euros. This does not yet include loss compensation, maintenance, repair or capital costs. The long-term average price of electricity on the stock exchange is five cents, just one percent of that. If we compare that with a gas storage facility in a salt cavern in Germany, we are worlds apart. Only around 50 cents need to be invested for one kilowatt hour of storage capacity in the cavern. Calculated over the same 20 years, each with one cycle, this results in 2.5 cents per kilowatt hour. In addition, there is the depreciation for the electrolyzer and gas turbine, but these are rather subordinate. That is perfectly okay for a long-term storage facility, which is a form of insurance cover. In this sample calculation you could also add five to ten cents per kilowatt hour for the losses that result from the poor efficiency of a hydrogen storage system, without that being a problem. Of course maintenance, repair or capital costs must also be taken into account here. 

For this reason it so important to make a clear distinction between the forms of storage: short-term storage must have high efficiency above all else, and long-term storage must have low investment costs as it is so rarely used.

In your view, are there any storage technologies that are currently under the radar but have greater potential?
I would say it is more the other way around. There are still many technologies that play a role in public opinion. However, it is quite clear that these technologies will never attain economic viability.

Unfortunately, it is very brutal: For storage in the stationary sector, it is virtually only about lifecycle costs. In the mobility sector, weight still plays a major role, as does performance in terms of super-fast charging. In contrast, the requirements we have for stationary storage are all harmless. The performances are manageable, which means we need maybe half an hour, maximum four hours of electricity from the storage. From a storage perspective, the maximum currents are small. Weight and volume play a very minor role. Therefore, in the storage sector, the qualification for stationary applications is almost exclusively based on the cost factor.

Redox flow batteries are discussed from time to time as possible alternatives. However, the problem is that lithium-ion batteries have a 30-year head start on the market. In 1991, the price was perhaps $3,000 per kilowatt hour; today, the price paid by vehicle manufacturers is perhaps $100 per kilowatt hour. In other words, a factor of 30 has been gained here, largely through economies of scale in production. It is virtually impossible to catch up.

There are always discussions about technologies such as flywheels, double-layer capacitors or superconducting coils for even shorter hold-up times. However, we find that they are not economically competitive – and that is always the essential factor. In other words, technically the systems all work, but over operating times of 15 minutes or longer, these systems are simply not economical. For this reason we really only see them in special applications and less for the system integration of renewables. An example would be a container crane that lifts a container every minute and puts it down again and has a high peak power demand for this or a regenerative power demand through recuperation. A similar application exists, for example, in the substations of subways or tramways. In all cases, peak load capping is the motivation and not the integration of renewable energies.

In your view, will electric cars also play a noticeable role in balancing out fluctuations in demand within a day?
If I just take the smallest wallbox with three kilowatts as the charging station and assume 40 million cars, I get 120 gigawatts of power. If you compare that with the six gigawatts we have in pumped storage power plants or the 80 gigawatts of peak power, you can immediately see where the potential is. And it's absolutely clear: vehicle batteries don't get broken from driving, but from standing around. If we don't succeed in utilizing this huge amount of investment in the field of electric cars, it would be extremely counterproductive. It would mean that we would have to build additional storage facilities, which would cost additional money and make electricity and energy more expensive for all of us. 

I not only see this as a great opportunity, but even as a necessity. On average, a vehicle runs for one and a half hours a day. Therefore, there are huge capacities available most of the time.

With regard to the marketability of storage technologies, what measures does the new federal government need to take here?             

One very important question is: how do we get stability into the distribution network? We are talking about the medium and low-voltage range here. The regulation we have today actually only ever leads to more power lines being laid when there are problems in a distribution network. What has not been regulated at all so far are the possible alternatives. The grid operator could –  instead of expanding the grid – also make contracts with those who have electric cars, with those who have electric-based heating systems, home storage and photovoltaic systems. So it's a matter of buying in flexibility and, therefore, the ability to regulate. 

At the same time, there is another significant aspect at play here. Up to now, grid stability has essentially been ensured by the transmission system operators via large power plants. In the meantime, the question arises: can't this control power actually be obtained from the distribution grid, for example via a collective of millions of electric vehicles? In my view, there is absolutely no reason why instantaneous or primary control power should still be made available and traded in any special units in the future. In fact, these distributed systems can do that extremely well and extremely quickly. The provision of, say, one or two percent of the connected load as frequency-dependent control power could simply be part of the grid connection conditions. In principle, they can be at full load in ten milliseconds with any battery via the power electronics. For stability, it is much better to have distributed storage than to have the few large pumped-storage power plants today. These are only located at a few nodes and do not solve problems that occur in the distribution grid at all.

Another aspect that plays a role is the issue of network charges. In industry, we have a commodity price and a demand price for electricity – and that is also appropriate. We don't have that at the household level. We only pay a commodity price there. But I think that's where we actually have to go. For example, if you want a very powerful charging station for your electric car and think you have to charge your car at home with 22 kilowatts, then you should have it and pay for it. And not just the kilowatt hour, but also the network load. So if you introduce a capacity charge, then it also makes economic sense to smoothen peak loads. That relieves the load on the power grid and avoids expansion and is a business model for storage.

And one final point which is important in terms of regulation: we have to get to the point where we price electricity dynamically. If we have surpluses from wind power in northern Germany today and the plants are shut down, that doesn't make much sense. If there were a second electricity tariff for surplus electricity, then, for example, in all houses that have gas heating, the immersion heater in the water tank could be started, the gas burner switched off and fossil gas saved. In this way, electricity is used locally. However, nobody will do this as long as the electricity costs 30 cents for the end customer. Here, too, a radical fresh start is needed in the question of how this energy market as a whole is controlled and set up.

Have these tasks already reached the political arena?
I believe that the parties in the "traffic light coalition" are aware of the issues. They know that overall we are at the end of the road with the energy market design. Even the merit order principle that we have for the power exchange no longer works in a system in which we have a lot of feed-in producers who have marginal costs of zero. The principle that the last marginal producer determines the overall price of electricity cannot work in a world in which mainly plants with zero marginal costs are operated. It simply has to be said that the last government and the last Minister for Energy and the Economy did not want to tackle this.

Of course, there are always a few people who might have to pay more after a reform – that is the big dilemma that politicians as a whole face when they want to make such reforms. Politicians are very worried that a few losers will be so loud in today's media world that ultimately everyone else will think that what has been done is bad. Unfortunately, this fear strongly guides the actions of politicians. 

If we look ahead to 2030, which storage technologies do you think will dominate which applications?
In the battery sector, it will continue to be lithium-ion batteries. Maybe lithium will be replaced by sodium in some places, but from the user's point of view, that does not play a role for stationary storage. What I mean to say is that certain individual details may surely change: how much cobalt or how much nickel or whether it's more iron phosphate or sodium instead of lithium, or solid-state electrolytes instead of liquid organic solvents, there will certainly be some developments here. However, that will come organically, without making a revolutionary leap from the user's point of view or having to change the basic design of storage systems. In any case, the basic technology is already clear today. For users, the only thing that matters in the end is which change can lead to a reduction in costs.

And then it will be a matter of needing gas storage. Technically, this is relatively simple. Electrolyzer technology for hydrogen production has been used for more than 40 years. There, too, it's a matter of making it cheaper, more durable, more efficient. However, in principle, the technology is available. For long-term storage, the main issue will be market design. Personally, I don't see reserve storage in the normal electricity market. If energy is to be held in reserve for dark and wind scarce periods, it may be that there is no demand at all for the larger part of the storage for several years. However, at some point, the three-week dark and wind scarce periods will come. That is the classic insurance principle. In my opinion, this should be solved via capacity markets. The storage facility would then be treated like the grid infrastructure: the state or the Federal Network Agency orders corresponding capacity. It's just a question of when the state starts to put these capacity markets out to tender. Building the storage facilities is not witchcraft, especially since we don't even need major planning approval procedures here like we do for overland lines. That is why it can be done relatively quickly. The gas storage facilities will be built exactly when the government decides how much reserve capacity we need. Under the current structure, that would be the job of the Federal Network Agency. Today, it solves the problem by moving coal-fired power plants that are actually already out of the system into the cold reserve. The only question here is when to make the switch.

Will we be able to provide the required storage capacities in Germany alone, or will we need – even more than today – large storage basins in Scandinavia and/or the Alpine region? 
The hydropower plants in the Alpine region do not even have approximate capacities for three weeks. The pumped storage power plants currently operating in Germany have an average of eight hours of reserve, the Swiss and Austrian power plants have a little more, but we don't consider the long-term reserve in the Alpine region at all. There are some wild ideas for the Scandinavian region. There are fjords which are 1,000 meters deep. If you build a dam there, one of the very large fjords would be enough to act as a "battery storage" for Europe. However, apart from issues like resettlement and massive ecological interventions, I don't think that is the solution. You would also need extreme line capacities. You would have to cover 80 gigawatts of peak load with lines, which would then only be used very rarely. I really don't see that happening.
In addition, Germany is, for once, very well supplied with mineral resources. We have sufficient cavern storage facilities. Today, we have natural gas storage facilities for 90 to 100 days, and we have further capacities that we could expand. For Germany, these salt domes are actually a mineral treasure that we have in almost unlimited quantities. The question of how to fill the salt dome is a completely different and exciting question. Is it hydrogen produced from our own surpluses, or is it hydrogen imported by ship or pipeline. But I think we also agree that the amount of wind and photovoltaic power has to be massively expanded, and with that automatically come surpluses that can only be put to good use in hydrogen. So hydrogen will also be produced in Germany. The question is, what will it be used for? The demand for hydrogen is much greater than the amount that can really be reasonably made available  – at least for the next 10 to 20 years. 

There is another aspect to long-term storage, and that is the resilience of the system. If you are only connected by power lines, you have an extremely vulnerable system. If someone disconnects the power cable, then within seconds the entire power supply fails. This is much less critical in gas systems because the time component is significant. If Russia turns off the gas tap, it takes five days before we feel it, and then we still have our vast storage facilities.

I cannot imagine building our energy supply on thousands of kilometers of cable to outside our EU partner countries, which cannot be protected at all. This is a different story within the EU, for example to Scandinavia, Italy or Spain. Here, line expansion should have a high priority. Germany will remain an energy importing country. I am convinced that we will import energy from outside Europe via hydrogen or via chemical energy carriers produced from green hydrogen. However, this is much less critical in terms of reliability. This should be a high priority for economic policy reasons. If Europe needs more development in the south, then we can all benefit from it. Italy and Spain, in particular, are greatly increasing solar energy, and we are buying the energy. 

About Dr. Dirk Uwe Sauer: Dr. Dirk Uwe Sauer is Professor of "Electrochemical Energy Storage and Storage Systems Technology" at RWTH Aachen University. The physicist, who holds a doctorate, deals with all facets of battery and energy storage systems for mobile and stationary applications. In addition, Professor Sauer is, among other things, also the leading director of the policy advisory project of the national science academies "Energy Systems of the Future" and represents the topic of energy in the presidium of the German Academy of Science and Engineering acatech.