How a Fraction of a Battery Can Cripple iPod Production


This week, the Wall Street Journal (s nws) reported on a possible delay in iPod (s AAPL) production because of the shuttering of a polymer manufacturing facility, owned by Kureha Corp., near the Japanese earthquake’s epicenter.  As the WSJ reported, the polymer in question is called polyvinylidene fluoride, or PVDF.  PVDF is used as a binder in lithium-ion (Li-ion) batteries. The WSJ says Kureha owns 70 percent of the PVDF market. (This surprised me, but I trust the WSJ.)

First some background. Li-ion batteries are made by taking powders of the active materials (which hold the charge) and using them to coat a conductive current collector to make an electrode.  The procedure to make the electrode is similar for both the anode and the cathode (or more accurately, the negative and positive electrodes).

Once you make the electrodes, you take the anode and cathode and put a layer of separator between them and wind them into a “jelly roll”.  You can also stack the anode, separator, and cathode layer-by-layer.

So how do you take a bunch of powders and make them stick to a current collector?  Well, you bind them using a polymeric binder.  First you make a slurry by taking the active particles, the binder, and other conductive stuff and mixing them up with a solvent. Then you cast the slurry onto a current collector and heat it up to evaporate the solvent. You are then left with the active particles and the conductive stuff held together by the polymer binder.

And you guessed it: The polymer is PVDF.  If you don’t have the binder, the electrode would just be a bunch of powders.

The Importance of PVDF

So how important is the binder?  To borrow a page from MasterCard:

  • The percentage of Li-ion batteries that use a binder: 100 percent.
  • The percentage of the electrode’s weight the binder takes up: 2-6 percent.
  • The number of different binders that can be used:  1.
  • The look on the production manager’s face of the battery company that uses Kureha PVDF upon hearing of the plant closure: horror?

Kureha PVDF is the one thing money can’t buy. For everything else… well, you know the drill.

Why not use some other polymer?  There are a few options (very few), but none work as well as PVDF.

So is there no other supplier of PVDF?  Turns out there are others (one U.S. entity comes to mind).  But changing suppliers in the battery industry is not the same as changing the supplier of flash memory for your iPad.

Why, you ask?  Well, when you change suppliers, your new material may be of a different purity, different molecular weight, different functionalization, and different particle size.  All this will have an effect on the quality of the electrode and the process used to make it.  Any company that decides to change suppliers will need to spend the time to qualify the new supplier and convince themselves the battery will work adequately.

In other words, it’s not so easy.  Chemistry has a nasty habit of being difficult that way.

So, if Kureha shuts a manufacturing facility down, all battery companies connected to Kureha are going to see delays (unless they have a stockpile). I’m not sure why the WSJ specifically noted the iPod was affected, as it seems to me that this would have a wider impact.

Lets hope that the problem with Kureha is overblown and that companies have enough stockpile to last them for a while.

The Side Story

The story doesn’t quite end there.  It turns out PVDF has other issues associated with it.

Notice that the PVDF is not actually active, but is taking up space and weight in the battery (albeit a small amount).

And it turns out that the solvent used to make the electrode slurry is N-Methylpyrrolidone (NMP).  NMP is pretty nasty stuff and not that great for the environment.  When you remove it in the drying step, you can’t just let it go in the effluent stream.  You have to capture it.  This adds cost.

The drying of the electrode itself has to be done slowly to prevent cracking while drying.  The slower the speed, the more the cost.

If NMP is bad, then why not use something else instead?  Because, PVDF dissolves in few solvents and NMP is one of them.  The other solvents aren’t quite as good in playing with the active particles.  Meaning: Using something else can kill your battery life.

Did I mention that chemistry has this nasty habit of being difficult?

Researchers have been trying to get away from PVDF and NMP and change to water-soluble binders.  These have had mixed success, but maybe in time something will change.  There has also been a lot of effort put into making electrodes using dry processing (i.e., not using a solvent). Again, this is still being developed.

The Big Picture

Let’s step back from this specific news item and examine the larger picture. To me, this incident highlights three things:

First is the issue many have worried about: The battery supply chain is in Asia with little done in the U.S.  It could be argued that it’s of strategic interest for the U.S. to not be dependent on the vagaries of nature (and, in some cases, the attitude of other countries toward the U.S.).

Second, it highlights the complexity of a battery: A very tiny component can have a big impact.  This issue isn’t specific to a battery (e.g., you may not want to buy a particular brand of car if the car company decided not to put in brake pads because their supplier was out!), but in many other industries, there are alternatives.

Here one supplier can have an impact. The same is true for the active material in the battery.  Change suppliers and you have to re-qualify everything, and the process is probably going to change a bit. This takes time. And time is money.

Third, it brings home how some processes in the battery (in this case the slurry coating and drying) are not that efficient and add cost.  Nobody wants them, but we haven’t yet found something better.

Some have argued that what we need are alternate processing methods that get away from some of these expensive steps and to find new ways to make electrodes that are fast and lower cost.  It’s hard to determine the exact cost to manufacture a battery, but it isn’t negligible (estimates suggest that it is  about 50 percent of the battery cost, with the rest going to materials).  Finding new ways to make a battery could make a cheaper battery, but also a battery with more energy. (And did I mention that batteries are expensive and don’t have enough energy density?)

This is easier said than done.

If we really want to do this, we need to find ways to get battery companies (who, in the U.S., are still in survival mode) to work with researchers in academia and national labs (who have the luxury of thinking a bit long-term). When something is two decades old, it’s going to be hard to replace it.  But that doesn’t mean we shouldn’t try.  To succeed, we need a comprehensive effort and some long-term vision.

In the meantime, maybe the U.S. needs to start hoarding PVDF?  A strategic PVDF reserve, anyone?

Venkat Srinivasan is a staff scientist at Lawrence Berkeley National Lab and writes about batteries on his site This Week In Batteries.

Image courtesy of tonystl.




I know you don’t like to comment on specific batteries, but is this a better battery or better battery hype.

A couple of topics I would be interested in seeing your comments on.

How do we sort through researcher/corporate hype?

Why is commercialization such a barrier to new battery/energy storage technology?


You said, “But changing suppliers in the battery industry is not the same as changing the supplier of flash memory for your iPad.”
I realize that that is the common understanding of flash, but– You should check with actual users (that is, phone manufacturers, for example) of flash memory before you make comments like that. As Eli Harari says, “Not all flash is equal.” Look at the market for flash on DRAMeXchange–you will see large differences between the high and the low prices paid, much larger than the differences between the high and low prices paid for DRAM. And the reason for that is that flash chips made by different vendors are not simply interchangeable.

I realize that this point is peripheral to your post, but I have seen this misunderstanding too often to just let it pass without comment. And I also second the comment from a previous poster about the WSJ–better second source its numbers. That goes not just for them, but for all newspapers. And magazines. And, for that matter, even me.

Venkat Srinivasan

Point taken. Did not realize the issues with flash.

I heard from a source that he thought that the 70% number may be for a specific formulation of PVDF and not for the whole market. Need to check into this further.


“…but I trust the WSJ.”

I would definitely find a second, reliable source for anything I read in the WSJ. It’s not the organization it once was.


The Japanese don’t like to share their secret sauce. Greed and paranoia left them to build manufacturing plants on active fault lines. But the US isn’t any better, Silicon Valley sits on a major fault line and when the Big One hits, a substantial portion of the tech industry will grind to a halt.


Look for this,an ipad alternative:


Another big picture thing that concerns me is the number of component manufacturing facilities affected. It blows my mind that these global companies would have key manufacturing facilities in an area that is extremely prone to earthquakes. Further, you discuss the environmental impact the components can have and again it was someones idea to put these components on a fault line? Yikes.

Comments are closed.