Engineers look under the hood of electric car batteries

We’re going to examine the EV market - which is currently dominated by Elon Musk’s Tesla and the Chinese company BYD. Both manufacturers have released limited data on the batteries that sit at the heart of their technology, so the structure, characteristics and performance remain shrouded in mystery. And there’s nothing engineers like more than a mystery to unpick! Which is why a group of researchers have been buying up and taking apart these batteries to find out how they work! We asked Rhodri Jervis - an expert on batteries at University College London - to take a look under the hood and guide us through what the paper just published on the work has revealed…

Rhodri - This paper is a type of paper we call a teardown analysis of commercially produced batteries by some major battery manufacturers. And the point of the paper really, and papers like this, is for the researchers to do a comprehensive analysis of what goes into the battery, how the battery is made and how the batteries perform. When you buy a battery commercially, it often comes with a data sheet, but that data sheet can be quite basic really, so basic information about things like voltage limits, temperature, operation limits, capacity, how quickly they can charge and discharge. Maybe if you're lucky, they might tell you what chemistry is actually in the battery, but not much beyond that. And so these kinds of papers can be really useful for researchers such as myself to be able to understand what's going on in the battery, particularly for people doing modelling work, because they can use all of these numbers and stats to parameterise their models.

But also I think of great interest to the public as people are getting more and more conscious about batteries. I think these kinds of papers, though they are obviously written in an academic language, they can provide some real insight into the incredible amounts of engineering and material science that goes into lithium ion batteries working as they do.

Chris - Can we actually learn a huge amount though? Do the manufacturers not try and protect their tech by making it hard to tear these things apart in a meaningful way? Do you not destroy it in the process and then that negates your learning? Or actually, are they quite easy to pick apart, see how they work and then copy them?

Rhodri - They are very hard to take apart and one should know what one's doing to do that. So, I'd recommend that people do not do this at home. But the most important thing is to obviously completely discharge the batteries before we open them. We open batteries in our lab fairly routinely, but there are very strict safety protocols to do that and for good reason. Lithium ion batteries are very energy dense and they have some materials in them that might be toxic and whatnot. So I would implore people not to take the batteries apart themselves.

And because of that, I guess people see batteries as a bit of a black box, you know, it's something that you don't really get to see inside all that often. And so this kind of work can be really enlightening in that way. You can tell an awful lot from these types of teardown analysis. The manufacturers might keep the exact chemistry of the battery somewhat secret, but it's quite easy to tell what that is once you've opened up the battery, as long as you've got the kind of analytical equipment that we have in universities.

But there are some secrets that are very hard to know, even with a well equipped lab taking these batteries apart. And one of the key ones is the special formulations that they use in the electrolyte. This is the liquid that is in the battery that kind of helps the lithium ions travel from the positive to negative terminals. There are lots of special things called additives put in these batteries. And a lot of these additives actually get decomposed in the first few cycles of the battery's life. And those cycles will happen in the factory when they're made. They're called formation cycles. And that forms a really important, very, very thin protective layer on the surface of the graphite particles on the negative electrode. That's something called an SEI, and it's really critical to how the batteries perform over a long time.

And because the additives that manufacturers put in those electrolytes tend to be pretty much consumed in that process, it's very hard to reverse engineer that. So, I'd say that's the area that's the hardest to understand and probably the closest guarded secret in the industry. But things like the structure of the electrodes, how the cells are built and put together, and even the types of materials that are in the positive electrode where there's more choice on those materials, that's all possible to understand from this kind of paper.

Chris - It's a bit, I suppose, like microchips. We know how microchips work, but to actually make them, to have the machine capable of operating at the resolution it does, to etch silicon the way it does, that's where the real money is, isn't it? So, although you can pick the battery apart and say, ‘oh yes, I can see how that works’, there must be elements to the engineering and the manufacture that are much harder to unpick or reverse decode or decipher from what you see in front of you, and that's where the IP is.

Rhodri - Yeah, I think that's true. The batteries operate on multiple different length scales, right? There are important things happening right at the atomic scale. And maybe one of the key things that the contrast of the two batteries in this paper is how important actually the sort of macro scale engineering is in the batteries as well, how you put all of these things together in a particular can or whatever it is. In this paper they compare two really quite different batteries actually.

The Tesla 4680 is what we'd call a cylindrical cell, quite a large cylindrical cell, and inside that all of the electrodes are rolled up in a spiral that the Americans call a jelly roll, that you might think of as more like a Swiss roll.

And then the other cell by BYD is called a blade cell, and this is nearly a metre in length and with a kind of flat form factor, so very relatively thin, really long. And that's the kind of cell that you might have in a laptop or a phone, but just on a bigger scale. And the different decisions that you make for those different form factors of battery really are quite key, and there's reasons for doing both.

Chris - What did they actually find? When you read through the paper, what leapt out at you as, ‘well, that's a surprise or that's interesting’? What were the main findings about this?

Rhodri - I think some of the approaches that have been taken in both of these cells were relatively well known to people in the field, I suppose. Particularly the Tesla cell, there was an announcement a few years ago about how they were going about making these particular cells, various different things like the so-called tabless design, for example, were known about. But I think this paper does a really good job of going into forensic detail in that, and so providing parameters and numbers that are really useful for modelling and also confirming some things that perhaps were suspected but maybe not known in detail.
So we get a range of things from photographs of the internal structure and components to chemical analysis of those components, electrical and thermal analysis of their performance when they're charging and discharging, and basically a full parameterisation of the cell. As I said earlier, some of the electrolyte components are unknowable because they get consumed, but other components do stay and they've done a chemical analysis of those as well.