Sunday, October 16, 2016

Finally, we know why the Note 7 exploded.

The answer, according to the New York Times:  Samsung seems to have packed it with so much innovation it became uncontrollable”.  Apparently phone components are like 4-year olds; put 3-4 of them in a room and they are ok, but pack 10 of them in a small place they will destroy the room with all the energy feeding off of each other.  Of all the theories, this one takes the cake. 

As much as I think the community should not be speculating on the underlying case of the explosion, I think we can’t help ourselves.  For battery types, this is the most interesting event that happened since the-last-time-there-was-a-fire, so we are all salivating.  What can I say: we don’t get out much. 

For non-battery types, I can see their need to know if more phones are going to be exploding.  Only way to know is to understand what happened with the Note 7. 

I have been getting this question steadily the past few weeks, so I thought I would collect all the speculation in one single place.  I also provide my views on this.  So here goes:

1. The battery was overcharged (BMS failure, too high an upper cut off voltage, aliens, whatever) and hence the fire.  I think the eager ones among us, who do not believe in waiting for more information, speculatively stated this.  I think Samsung’s revelation that there was a manufacturing flaw negates this theory.

2. Samsung used a 6 um separator in the battery and this lead to defects when assembling with a thin separator. This in turn led to shorts and the explosions.  As I explain below, this may be part of the story.

3. The battery had a manufacturing defect where the anode and cathode did not line up correctly, leading to edge effects, lithium plating and shorting.  This seemed very likely until the New York Times article came out. 

While these three issues are obvious ones that most battery types would guess, the New York Times articles makes the point that after initially concluding that it was the battery, Samsung realized that it was not that simple.  The article claims that Samsung could not pinpoint the reason!  Hence the wacky statement about “uncontrolled innovation”.

Let us be clear: Samsung makes great batteries. ATL makes great batteries.  If this were an obvious issue, they would have caught it after the first few weeks.  There is something to be said about the argument that this is a more complicated problem.

Since the Times article came out, we have had three more, system level, theories that have popped up.

4. The battery was being fast charged due to a chip design flaw.  Faster than it was rated for. This lead to overheating, thermal runaway, ending with you-know-what.  I’m not so sure about this.  The Note 7 phones were exploding even when not fast charged so…

5. There was so much being packed in a small volume that the battery was getting squeezed and the edges pinched, unintentionally, leading to shorts.  This theory does seem possible, but I like the theory below (which is a variant) the most. 

6. Samsung used higher content of silicon in the graphite-silicon blended anode.  The silicon expands on charge and swells the pouch.  Because the pouch was unable to swell in the phone due to the lack of space, it shorted and exploded.  I really like this theory.  There have been problems with battery swelling even before silicon came to the scene, but this has only gotten worse with silicon-graphite anodes.  Many consumer electronics companies have been worried about this and have a spec. for how much the battery can expand.  According to Mashable, the Note 7 had a 750 Wh/l battery; which is PRETTY energy dense.  Much more so than previous generations of batteries, suggesting higher silicon content than before. 

It is possible that higher silicon content combined with a thinner separator and less space in the phone for volume expansion all came together to lead to shorts and fires. 

I’m sure I’ve missed a few other theories (aliens?), but I think I got the majority of the ones I have heard.  

Now that we have that out of the way, let us talk a bit about what this all means. 

I think the initial speculation that this was a battery-level issue appears too simplistic.  Clearly, there is more to the story.  Batteries all over are safe, have been safe, and will continue to be safe.  Assuming you know what you are doing.

But what if you have “uncontrolled innovation” happen again? What is a poor battery to do if the overall system does not want to treat it kindly?  Li-ion batteries are energy storage devices.  Meaning, if you release the energy very very fast it is not going to be pretty.  So, some TLC is in order. 

But even if the system screws up how it handles the battery, shouldn’t the storage device itself be made to withstand any abuse?  As we move toward wearable technology with things attached to every part of our body, we need to ensure that the battery remains robust even if there is a system-level failure.

There has been a narrative going around that Li-ion batteries today are similar to where crystalline silicon solar cells where a decade ago: meaning, the prices are dropping and one can get installers and system integrators to come in and start to make them ubiquitous.  The Note 7 incident shows the perils of this thinking. 

Batteries are not plug and play devices where Jane-solarinstaller is going to buy something off of Alibaba and dump it in your garage and get you city permit folks to sign off as if they are inspecting your plumbing.  We better be buying from someone who knows how to make them well.  And we better know how to design the system correctly, install them well, and control them through the life of the device.

But to get the world where we do treat our batteries like we treat our microwave (bang on it to try to get it working?), we need the batteries to be robust inside out.  This requires a separate blog post, hopefully, in the near future.  

In the meantime, with the Note 7 off the streets, time for the battery folks to crawl back to the cave and focus on achieving a few breakthroughs.  Until the next incident… 


Sunday, September 11, 2016

Why are Note 7 batteries exploding?

Samsung Galaxy Note 7’s are exploding when they are charged.  The tech media is, rightly, trying to get to the bottom of this. 

Much has been said about how this can happen and the focus appears to be on overcharging Li-ion batteries and the resulting heat generation.  The only party that actually knows something about this is Samsung and they have said that there was a “minor flaw in the battery manufacturing process”. 

This does NOT sound like an overcharging problem.  This could be either misalignment of electrodes leading to lithium plating and shorting.  Or something like copper dissolution and subsequent deposition leading to shorts. 

Or something else. 

But more importantly, except for Samsung, we are all guessing.  As long as Samsung does not tell us (and I’ll go on a limb and say that they will not), we can make up anything we want.  But guess what: we are making stuff up!  We should probably stop. 

But where is the fun in that.  So let the speculation continue. 

Interestingly, the Chinese made batteries by ATL are not exploding.  Only the Samsung made ones are. So much for "low quality Chinese manufacturing"…


Monday, February 1, 2016

A Bolt (with a B not a V) from the Blue

From time to time, I have been known to get all excited trying to explain to my 3-year old that he is living in an amazing time.  Self-driving cars, hoverboards, home automation, robot assistants… the list of amazing changes goes on and on.  He, in turn, has been known to give me the “leave me the @$%#$ alone” look while he gets back to being amazed at the number of ABCD songs on YouTube. 

Apparently, both my son and I are wrong to be amazed. Atleast this is what I learnt when I read reviews of a new book titled “The rise and fall of American growth” by Robert Gordon.  His TED talk argues that the pace of technological innovation in the last 50 years is far lower than the period in the first half of the 20th century.  Unlimited ABCD is all nice, but from an economic growth perspective, apparently not as big a deal as indoor plumbing (my 3-year old just asked me to use said indoor plumbing to evict myself from the house when I pointed out to him that he is doing nothing of economic value to society).

I am no economist.  But I do know a little something about batteries.  And I have been known to say stuff like “there is nothing new in batteries” every time a paper comes out claiming a breakthrough.  I was mostly right for a decade.  Over the last 2 years, I have stopped saying that and have gotten to be amazed at the changes in the battery world.

No. No.  I’m not talking about the press releases claiming breakthrough.  Those are still not worth the bytes they are written on.  I’m talking about real changes, like the dropping prices for Li-ion batteries. 

Or the fact that I have stopped counting the number of EVs I see in my (rather long) daily commute.  From Nissan’s to BMW’s to Merc’s to Tesla’s, they are pretty much part of the landscape.  Granted, I do live in a bubble (the Northern CA area is not exactly representative of the world-at-large), but it is fascinating to see the penetration of these cars.

But as a low-paid National Lab scientist with a long commute, I have been unable to make the jump.  Either these cars had very little range (Berkeley, for all its greenness, does not excel at providing charging) or cost too much.  And gushing over a car at the dealership does not lead to a price reduction.  Quite the opposite!

But prepare to be amazed even more. 

Chevy recently announced the 200 mile Chevy Bolt (not the same as the Volt.  They really should start calling the car “Chevy Bolt: with a B not a V”).   Available sometime next year, it will be a small to mid -sized 5 passenger crossover with reasonable trunk space.  And Tesla is said to be ready to announce the Model 3, another 200 mile “less frills” car for the rest of us.

Both the Bolt and the Model 3 are going to be in the mid to high $30k range.  With the tax rebates, we are talking under $30k.  Even I may be able to afford that!  And 200 miles may be enough to remove range anxiety for most of us. 

And, with all the auto companies starting to agree on putting the battery at the bottom, space for passengers and for the Costco runs appears to be in the offing. 

Now, 200 miles is no 350 or 400 miles (typical gasoline car range).  Nor is the price comparable to gas cars (I suspect that a gas equivalent of Bolt is probably a $20k car), and tax incentives are probably not to be depended on in the long run, but these $35k/200 mile EVs are going to be a game changer. 

And more automakers are promising to go after the 200-mile target.  There is a feeling that this is the sweet spot for the next push for electrification.   With falling Li-ion prices, these cars are going to get cheaper year after year.  I do think leasing is probably a better idea considering the rapid pace of changes, not just in batteries, but also in the level of sophistication in the cars.

Granted that there is probably little economic growth in going EV (I suppose you are substituting from one production type to the other. And some of the battery cost reduction is coming from automation), but you have to admit that from a pure technological standpoint it will be good to have something new for a change on the roads after a century of the same kind of drivetrain. 

Combine the EV revolution with ubiquitous connectivity and self-driving technology, we are going to have a LOT of time on our hands to think about doing something of economic value in the silent cone that is our battery-powered Uber.  And nothing like boredom to invent something new to keep ourselves from getting bored.  After all, isn’t necessity the mother of all inventions?  Hopefully what we invent will be of economic value so that we can prove Robert Gordon wrong.

In the meantime, continue to be amazed at the changes you see on the roads.  When my 3 year old went through a (mandatory?) car identification phase he has been known to say stuff like “Too many Toyota’s”.  Before long, if the trend continues, I think he will be saying “Too many Tesla’s”. 


Tuesday, January 12, 2016

Back by Popular Demand-More Intercalation Chemistry, This Time with Sodium!

Folks: Happy New Year to all.

As I was off on my merry vacation, I noticed that the popularity of this blog had doubled!  And everyone seemed to be reading the post by Marca Doeff, our guest blogger.  After a few weeks of soothing by battered ego, I decided that one data point does not a trend make.  So we are trying to get another data point today with Marca's next post.  enjoy


It’s Marca again, enjoying her highly lucrative new career as substitute blogger for TWIB! What should I do with all the money I rake in from this gig? Buy a Tesla?  Hmmm, maybe a new BMW i8-they sure are pretty, aren’t they? Should I get a blue one or a silver one? Decisions, decisions! [Editor: Dream on. Get to the point, will you?]

            Ahem, well, okay.  When last we left off, I was describing the concept of intercalation as it applies to batteries; i.e., when lithium ions insert into electrode materials. What I didn’t get around to saying is that lots of other things beside lithium ions can be intercalated into host structures-not only cations but also neutral species like molecules or polymer chains, and even sometimes anions (although, as far as I know, oxidative intercalation of anions only happens with graphite or disordered carbons with graphitic domains). Even before the lithium-ion battery was officially A Thing, people were having all sorts of fun sticking stuff between the layers of clays or graphite to make new materials with interesting properties (I even tried my hand at it Back In The Day). For the nerds among you (you know who you are!), there’s a classic paper by Mildred Dresselhaus (The Queen of Carbon Science!) called Intercalation Compounds of Graphite[1], which will tell you just about everything you need to know.

            But I digress! At this point, you, my faithful readers (both of you), are probably tapping your feet impatiently saying something like “Well, if intercalation isn’t just limited to lithium ions, couldn’t we base a dual-intercalation battery on something else, like maybe SODIUM?” Very good! You all get gold stars, my nerdy blog-reading friends!

            The sodium-ion battery or NIB, is a subject near and dear to my heart. You see, back when dinosaurs roamed the earth, in the early 90’s, when lithium-ion batteries were just getting started, I was playing around with this concept in the lab. Stan Whittingham had described not only reversible lithium insertion, but sodium insertion as well in his early papers on TiS2. Scientists like Keld West in Copenhagen, and T. Richard Jow at the U.S. Army Research Lab were also publishing work describing some intercalation compounds as possible cathodes for sodium-based batteries. So, I wasn’t necessarily the very first person to work on sodium intercalation, but I was definitely an early adopter.

            The mad rush by the research community towards lithium-ion batteries at that time left me practically all by my lonesome to work on NIBs instead. (What can I say? Either I’ve always been ahead of my time or I’m hopelessly out of touch, I still haven’t figured out which). While NIBs, in principle, operate much like lithium-ion batteries, there are some differences. One is that sodium doesn’t really insert into graphite (the favorite anode material for lithium-ion batteries), for reasons that are too complicated to go into here (read Millie’s paper if you want to find out why!). There is some reversible redox activity with disordered carbons, though, and I published an early paper on that. I also had a lot of fun working on some cathode materials, and even patented one with the nominal composition Na0.44MnO2. It has a very robust and cool-looking tunnel structure, which cycles sodium ions in and out very well-and lithium ions, too, if you ion exchange it and put it in a lithium cell.

Tunnel structure of Na0.44MnO2, a cathode material for sodium-ion batteries.

            The excitement around lithium-ion batteries soon swallowed me up along with practically everyone else in the battery world. Almost nothing happened for more than twenty years in NIB research. Then, starting a few years ago, the field started heating up again.

Web of Science search for papers containing the words “sodium ion battery”.

            What happened around 2012?  Well, for one thing, lithium-ion batteries had matured and material scientists were looking around for something new to do. The battery community was talking a lot about “Beyond Lithium Ion” (which really meant taking a new look at some old chemistries!) And there were screaming headlines like this one:

New York Times, Monday February 2, 2009.

            Now, are we really going to run out of lithium? Not likely, at least not in the long run. But when you read that the Tesla Gigafactory is projected to need almost half the world’s current annual production of lithium hydroxide, you gotta wonder if there might be problems with the supply chain, at least temporarily. It takes time to ramp up production, after all. It might be wise to have a “Plan B” battery-wise, and there’s tons more sodium in the world than lithium, TONS. It’s a lot cheaper, too, plus there are some other cost-saving benefits to sodium systems, like the fact that you can replace copper current collectors on the anode side with aluminum because sodium doesn’t alloy with aluminum, whereas lithium does.

            Because NIBs are so close conceptually to lithium-ion batteries, the development time should be shorter than that of other “Beyond Lithium Ion” systems. We can simply leverage all the engineering knowledge from the past twenty-five years work on lithium-ion batteries. All these considerations got some people excited enough to start companies based on NIBs, like Jerry Barker at Faradion. He’s targeting e-bikes. Then there is Jay Whitacre, founder of Aquion. Now there’s a unique concept-a sodium ion battery using an aqueous electrolyte, for grid storage! That system is potentially incredibly cheap, which is especially important for that application. One of the cathode materials that Aquion uses is my old friend, Na0.44MnO2.

All this new NIB activity means that my old sodium-ion battery papers, which languished for many years, have started heaping up citations, sort of like Sleeping Beauty. Sadly, I have to be contented with fame, not fortune, because the patents expired some time ago. But that’s okay, since I’m raking in the big bucks here at TWIB instead!

Marca Doeff

[1] Advances in Physics, 2002, Vol. 51, No. 1, 1-186