The Weakest Link – Aging Lithium-Ion Batteries

In electronics, batteries are always the weakest link. Many factors can influence how rechargeable batteries age. Their capacity is affected not just by the number of charging cycles, but also by environmental factors like temperature or by the behaviour of their users.

RealIZM has met Robert Hahn to speak about how lithium-ion batteries age and what Fraunhofer IZM is doing to research aging models and test the reliability of these indispensable components.

Let us start with the basics: How do lithium-ion batteries age?

Robert Hahn: The aging processes are the same for all batteries, whether you use them in a mobile phone or in an electric car. Here at Fraunhofer IZM, we have been part of a number of EU-funded projects on the sustainability of electronic products. In most cases, it is the batteries that are the weakest link, because electrochemical systems are simply not as stable in high temperatures, and they do not have the same longevity as other components in electronics. 

In terms of longevity and environmental friendliness, we naturally want lithium batteries to last as long as possible. This is also quite crucial for the user. In electric cars, for example, the batteries are about the most expensive parts, so it is naturally good for their owners if they last longer. We all know it from our phones: After two or three years, the batteries start to get weaker and weaker. 

It is not just the battery itself that determines how long it lasts, even though its structure and the materials used naturally play a significant role. Factors like how it is used, what the ambient temperature is, or how fast it is charged all have considerable influence on its life expectancy as well. This is all quite complex, so there is a great deal of research still to be done in this field. The smallest change to the battery materials, for example, can have a major impact on its aging behaviour. We can always start new investigations into how these complex mechanisms work. 

Whenever we develop or improve batteries in our micro battery prototyping line, we also have to look at how often they can be charged and discharged. This helps us judge whether the battery we created is a good battery. That’s why we have the knowhow to say something meaningful about aging processes, and that’s why we are involved in these projects. We have worked a lot on the aging of Li-ion batteries in the past, so the trajectory was clear for us.

To what extent is Fraunhofer IZM conducting research on battery reliability?

Robert Hahn: Reliability is definitely very important. In future, there will be miniaturized hearing aids that sit directly on the eardrum. That means that the batteries used for them will also have to be very small. We cannot expect the wearers to go to the doctor every few months to get their batteries replaced. We want batteries that will last a long time and that do not lose capacity so quickly. If a battery lasts twelve hours on a single charge at the beginning, it should not go down to eight hours after a few weeks. And there is the requirement that the batteries have high cycle stability, that is, how often you can charge and discharge them. 

The hearing aid, for example, would have to be charged once per day, so that would be 365 cycles per year. The optimum would be fewer charges, so that you do not have to charge your hearing aid every day. This could mean that it would last 4 years, with approximately 1,000 cycles. These are the same requirements that are placed on mobile phone batteries today. The standard is 500 cycles, and within these 500 cycles, the capacity may drop to 80%.

For electric cars, however, this would not be sufficient, because you want more cycles in this range. Their capacity should only go down by 20% over a range of 1,000 to 4,000 cycles. Of course, it is much more challenging to assure the performance of a battery pack that consists of hundreds of batteries compared to a single cell. But it is important that a rechargeable battery experiences only minimal aging in each cycle, no matter what size it is.

Another great application for batteries is storing power from solar energy. Many people now have solar panels on their roofs and maybe a battery pack down in the basement, so they can use their own solar power at night as well. In that case, you also want a few thousand cycles, and it all needs to pay off when you factor in depreciation, because these battery packs are quite expensive. In order to make producing your own electricity financially worthwhile, the whole thing has to last a long time. At the same time, there is also the question of general durability: Even with fewer cycles, batteries can get too warm and degrade. As a result, their capacity would also suffer after maybe a year.

You just mentioned two factors: the number of cycles and where batteries are stored. What other factors are there?

Robert Hahn: There are many different aging factors, but let me mention some of the most important ones: The number of cycles is the foremost factor. The second is temperature. If a battery is stored in too hot a place or used at too high a temperature, it will also age much faster. The next key factor is the current. You may want rapid charging and discharging, although the rate of discharge is dictated by the application and should usually not be that fast. But in electric cars, for example, you want an extremely fast charge, which again puts the battery under a lot of stress and ages it more quickly. 

There is another factor, and that is the voltage. If the battery is fully charged, then it has a high voltage and ages faster in turn. The user doesn’t have to worry so much about that, because it’s already considered by the system. If, for example, the electronics notice that the user has the device constantly connected to the charger, they will lower the voltage. Think of the solar power storage in the basement: The unit may be charged to maximum voltage and then the homeowner goes on vacation. The power is suddenly not needed anymore, and the electronics have to regulate this. 

But it is not just the electronics that take these factors into account automatically. The users themselves can also decide what is important to them, because the electronics can’t know everything either. It’s generally possible with notebooks or perhaps also with phones: You can configure the device to work for as long as possible between charges. Or you have the situation where that is not as important, because you always have access to a charger. In that case, the electronics does not always have to charge the battery completely, which will make the battery last much longer. But the battery should also never be completely discharged, otherwise you would have so-called full cycles, and they are not good at all. The best thing would actually be for the battery to always be in the range between 30% and 70% charge. That’s the sweet spot where it lasts the longest. 

Some things are controlled by the electronics, some by the user. In some circumstances, you can’t influence anything at all. If the environment is too hot, then nothing can be done with small batteries. But with larger car batteries, for example, active cooling could be an option. 

And you test those things on batteries in order to predict how long they will last?

Robert Hahn: These are all factors that have an influence on battery designs. We can test some of them in our labs. We need the labs anyway for our own battery cell production and app development. We have hundreds of channels in place where we can always charge and discharge batteries. 

In projects on longevity, we take such batteries or battery packs and charge and discharge them again and again under different conditions. This enabled us to create aging models that show the dependence on various factors. Speaking theoretically: if you knew that the batteries would continue to be used in this manner and under these circumstances, it would be possible to predict exactly how long they would last until their capacity dropped to a certain level. 

In addition, consumer organizations like Stiftung Warentest want certain evaluation criteria to compare different products. They do not want the effort of charging and discharging the batteries thousands of times. Accordingly, it is possible to look at the electronics or talk to the manufacturers about whether the developed charger considers things like durability, or that it is not always charged to the maximum voltage, or that it is charged as slowly as possible. Perhaps the users can even configure this themselves. In this way, the knowledge gained can be used to ensure that the battery lasts as long as possible and that it can be conveniently replaced. 

So you do not only test batteries developed at Fraunhofer IZM. Who approaches you for these tests: other researchers or partners from industry?

Robert Hahn: Yes, both is possible. It is mainly smaller companies that cannot afford such battery test equipment themselves. Most lithium batteries are manufactured in China, and some of them are very cheap. A company might want to know if the cheaper battery is just as good as the more expensive one. We can test that here at Fraunhofer IZM. On the one hand, we can see whether the parameters on the label are actually met. In other words: Is the capacity really as specified? Our tests therefore help small companies in their selection of batteries. 

Then there are also public projects. These are usually funded by the EU or by the German Environment Agency. They consider things like whether it’s possible to give manufacturers some kind of specifications, such as how long their batteries have to last or something similar. 

Among these EU projects, we had one called “SustainablySMART“. It was about the sustainability of electronic products, i.e. how long these products actually last. Currently, there is another EU project called “PROMPT“. It’s about a new test programme for the longevity of consumer products. The programme can be applied by consumer organizations and market monitoring authorities across Europe. This development is supported with our tests. A lot of knowledge is already available, and it is possible to evaluate whether a product is a good product based on certain criteria and checklists. Another question was: Can the requirements be adapted by legislators, so that certain features, for example warranty or longevity, can be assumed? Of course this always includes the batteries.

So Fraunhofer IZM was involved in creating these checklists?

Robert Hahn: Yes, exactly. We also created our aging models as part of this project. The question was always whether it is possible to come up with a quick test that can still be used to calculate ahead and see which battery is better. Our conclusion, however, was that this is not as easy as you might think. The only thing we can achieve with this quick test is to filter out “black sheep”, that is, batteries that fall off right at the start. It is very complex and difficult to predict whether a battery will fail after 500 or 1,000 cycles without more precise and longer tests. There are so many subtleties, so that even batteries from the same manufacturer may behave differently under only slightly different conditions. The result was that this is essentially not predictable. Of course, this is also a kind of result. Not everything is always as you would wish. 

Has this project already been completed?

Robert Hahn: No, the PROMPT project is still up and running in the capable hands of Anton Berwald here at Fraunhofer IZM. Our part in the project concerns only the batteries. Other departments investigated other electronic components. How long electronic circuit board lasts, for example, right through to washing machines or other consumer products. Basically, it’s about: How can you tell if something is high quality or not? 

In the case of cordless screwdrivers, for example, we examined how long the batteries last. Stiftung Warentest was also involved in the project and also conducted tests. I said earlier that the battery is usually the weakest link. But they completely tested these power tools by screwing very heavy screws in and out. This really resulted in the maximum possible mechanical load. Some devices simply fell apart, but the battery still worked in the end. In that case, it was the mechanics that were the weakest link and not the battery. 

The Interview was conducted and edited by Jacqueline Kamp

Picture: Fraunhofer IZM