there is no better battery


Ryan and I have been going back and forth in his comments about the likelihood of carsharing services going electric. I think it’s unlikely because they’d have to spend too much time charging; he thinks they’re a good candidate for early rollout of the charging infrastructure necessary for such a switch. Most recently he said:

In practice, this could be achieved incrementally. Tweak business models over a ten year period through which you slowly switch from gas engines to plug-in hybrids to all electric, over which period, presumably, battery technology slowly improves. Needn’t be done all at once.

I think this is a point that’s worth making here and at some length: “presum[ing that] battery technology improves” is setting yourself up for failure.

In truth, there have only been a few noteworthy improvements in battery tech during Ryan and my lifetimes: longer-lived NiCd and NiMH batteries; some improvement in alkaline batteries; and the popularization of lithium batteries. But look closer and you’ll realize that most of these aren’t actually battery innovations, per se: they’re benefits of the microprocessor revolution. Cheap, smart charging circuitry allowed us to avoid memory effects; to balance load across cells; and to monitor lithium cells’ temperature and voltage as they charge so that they don’t catch fire (well… usually), thereby finally making lithium a viable option for consumer electronics. Those are all important developments, but at this point we’ve wrung about as much as we can out of charging our batteries more cleverly.

None of this has done much to improve the fundamental energy storage densities of the underlying chemistries. These have been known for a long time now, and nothing is going to change them — nor are there any more promising elements like lithium waiting to be tamed (well, none that aren’t radioactive, anyway). The glacial pace of improvement in battery technology really can’t be overemphasized. The lead-acid battery was developed in 1859, for pete’s sake. It’s really heavy relative to the energy it stores, can produce explosive fumes if overcharged, and sometimes requires the addition of distilled water. Yet it’s still the best battery technology we have for supplying the high current necessary to turn over an engine. A century and a half and we haven’t come up with anything better!

It may seem like batteries have improved dramatically — consider the lifespan of an iPod Nano versus a portable cassette player. But this is misleading. In fact it’s a byproduct of more energy-efficient technologies. Which isn’t to dismiss energy effiency! But electric motors are already extremely efficient. And when it comes to vehicles, we’re unfortunately dealing with hard physical limits related to how much energy it takes to move a car. So long as we’re committed to EVs being able to perform like and drive safely near gasoline-powered cars, we will find ourselves with less room for improvement than people would like to think.

I don’t mean to be a downer, but it’s difficult to overstate what a serious problem this is, or for how long it’s been one. Hydrocarbons are an unbelievably efficient way to store energy when compared to electrochemical cells, and I seriously doubt anything will change that. Hopefully I’ll be proven wrong. But smart people have been working on the battery problem for decades and decades, propelled by the lure of the financial bonanza that a breakthrough would represent. And while they’ve made impressive improvements, none come anywhere close to competing with gasoline’s energy density. We’re still an order of magnitude away.

Now of course there are always fuel cells. And nanotech’s vast surface areas may deliver unexpected breakthroughs. But a bet that counts on a better battery is still a very, very bad wager.

About the author

Tom Lee


  • The problem isn’t with carsharing companies, it is, as you indicate, with the state of electric cars. It will be very interesting to see the Chevy Volt and the EV(s) that Nissan says it will be selling in the US in 2010.

  • Why such a pessimist? Battery technology has indeed been slow, but holding more electrons in a given space hardly seems like an insurmountable problem.
    It seems like fuel cells would only work if we could figure out a way to use nanotechnology to make complex hydrocarbons. There is no way to fit enough single hydrogen atoms on board a car to make fuel cells a reality. A gas just takes up too much space.

  • I’ve written before that the future, it seems to me, is not better batteries but safely electrified roadways that power a car as they drive. Your car would have some sort of transponder that recorded your energy use and you’d be charged like a utility. Would it be a major technological and infrastructural undertaking? Yes. But so is any solution to our growing energy problem, global warming or no.

  • That’s a beautiful dream, Freddie, but I doubt the infrastructure costs would ever work out. It would be easier to subsidize the build-out of a rail network that could transport the cars between cities. Tearing up that much road and installing wires under it — it’s just not feasible. And then there are the transmission losses to consider…
    For travel within cities, though, I can imagine such a system might work. San Francisco and some other cities use trolley buses that are more or less what you’re describing (some cities used to use underground vaults with electrified cables, but it was unreliable and extremely expensive). Overhead wires can probably be used for more applications than they currently are. Certainly it’d be great to see more trolley buses around American cities.

  • Thanks for your perspective. You explain this very well. There are many naive people and organizations that don’t understand the underlying limitations of batteries. Examples-
    **The US military offers a prize for an energy storage device that can achive an energy density slightly better than current theoretical Lithium limits. The prize is 1 or 2 million dollars.
    **Formula 1 car racing will allow kenetic energy recovery devices on the cars next year (2009). These will allow energy recovery during braking and energy output during acceleration, fairly typical hybrid-type stuff. The specific rules limit the system mass (~60 kg), the rate of power in/out (~60 hp), and the max time per lap you are allowed to use the output (~12 seconds). These parameters are already extremely difficult to achieve, and may not be possible at all. However, the Formula 1 rule makers happily predict they will simply increase the maximum allowable power to 120 and then 240 hp over the next few years, thereby forcing an improvement in the technology.
    There are many other examples with the common thread: People realize better batteries or battery-like ability would be great for their field, they do not understand that batteries are already approaching the energy density limit, and they assume small incentives like a cash prize or slightly quicker laptimes will somehow create new technology. Then they think their field will kindly bestow this improvement on the rest of humanity because nobody else was clever enough to create the incentives they did.
    I think you left one good point out of your article. There is already a tremendous incentive to find a better battery. A significant improvement in the best battery energy density (say double) at no extra cost would create a human improvement comparable to curing cancer or creating cold fusion. Any person or organization that could do this would simply become the richest person or business in the world.
    If people understood there was already a huge incentive to improve batteries and there was no resulting battery progress, then they would better understand that batteries have reached their limit.
    Thanks for putting up with the long rant of an anonymous person.

  • Thanks for the kind words, Fred. I agree completely that the usefulness of battery-related prizes or non-market incentives is regularly overstated. I didn’t discuss it here because I did so earlier — I probably should’ve included a link to that post in this entry.

  • Batteries can oxidize a wide range of fuels because they retain the ash. It is possible for internal combustion systems to do this too by a judicious choice of fuel. If the ash can leave the combustor, but then be cached in a compact cache in a waste cache bin, it can work better than a battery.

  • I disagree with the assessment that sufficiently good batteries don’t exist. The key question is energy density – the amount of kw.hours you can store in a kilogram.
    Modern batteries have hit the densities required. You need about 20 kw of output for four hours to make a car with a normal range. That’s an 80 battery. The existing Tesla Roadster has a smaller pack, 53 kw.hrs.
    A LiIon battery has a density of 0.16 kw.hrs/kg so your battery for the car above has to weigh 500 kg. That’s heavy, but not insurmountably so, once you get rid of a gas engine and gas tank.
    If you are willing to have a commuter car with half the range of a normal car (recharge it every night), you can halve the weight and expense of the batteries. This would be fine as a second car.
    The lithium sulfur battery is projected to have an energy density two to four times higher, so the battery mass will be just a couple of hundred of kg for a normal car. Just 15% of the car’s mass would be battery.
    The issue is battery expense, not the technology. Make existing batteries cheaper, and electric cars will be better than present day gas cars.

  • But we don’t really need the batteries to be better, we need them to be cheaper. So we don’t need better battery technology, we need better battery-manufacturing technology.

  • You’re right about theoretical limits on batteries. The chemistry involved has been quite well known for a long time. But you’re neglecting “other” technologies than purely chemical.
    I remember reading about a guy who was storing energy in a rotating electromagnetic flywheel. Put energy in by speeding up the flywheel, and take energy out by turning it into a generator. The main problem is finding a suitable flywheel that doesn’t explode at the tremendous speeds it was spinning at.
    Or how about the ultra-capacitors that I read about that MIT was working on? Takes seconds to charge and can hold energy at densities greater by several magnitudes of conventional batteries.
    Or even the zinc-air batteries? Powers a laptop for twenty hours in one charge.
    The point is this; just because we believe one avenue of research won’t prove fruitful just because twenty or thirty years ago researchers said it wouldn’t work doesn’t mean further down the road we won’t be able to overcome that stumbling block.

  • Rosco: I would suggest you remain skeptical of any battery technology that’s promising a doubling or quadrupling of performance. I hope it works out, but am not particularly optimistic.
    Johann and Goishin: supercaps are cool, but so far they can’t store nearly as much energy as an electrochemical cell of the same size. Maybe that’ll change, but again, it would be a mistake to assume that it will.
    As for flywheels — yes, they’re great for storing energy. They have to be extremely massive to do so, though, which means it takes a lot of energy to move them around. And what happens in a crash? Imagine a cascading failure of massive flywheels hurtling down the freeway, setting each other off. It would be horrific. Flywheels aren’t really viable for vehicles smaller than a tank. Finally, Zn/air isn’t a rechargeable chemistry (I think you’re also overstating its energy density considerably, most likely due to someone quoting an implausible energy consumption for the laptop you allude to).
    Josh: actually, it’s both price AND capability that need improving. Tesla clearly designed their roadster without much concern for cost, but the batteries they use still deliver a fairly anemic range. By emphasizing the car’s impressive acceleration, they seem to have managed to imply that electric cars can be awesome, it’s just that it costs too much to make them so. But designing an electric vehicle with great pickup is actually really easy — it’s range that’s the crucial problem. So it’s not just a question of cost. Cheaper batteries would be great, and would let us do more things. But we really do need energy storage technologies with greater capacity, too.

  • Fuel cells seem the most promising in the short term, the main problem is cost but someone just invented a fuel cell replacing the platinum with GoreTex, of all things.
    The guy with the flywheel design was building his flywheels out of carbon fiber. A failure resulted in the wheel bursting into small easily-contained fibers. Don’t know about the “bouncing down the road” problem but it seems likely that the wheels would burst if they hit the container walls.
    Ultimate solution is probably superconductors.
    Then again, if you use algae-based biofuel, internal combustion engines are just fine.

  • One thing I’ve not seen mentioned in this discussion is the possibility of swapping the batteries out – don’t wait for them to charge, just exchange the battery unit for a previously charged set. Could that be workable?

  • The key problem with battery energy density is that they have to store their oxidizer on board whereas combustion engines use atmospheric oxygen instead. Most promising avenues of battery research are in ways to safely and reliably use atmospheric oxygen in the battery.

  • Just FYI, car2go in Amsterdam is fully electric. They have a fleet of 300+ electric drive Smart cars (two-seaters) which are charged by charging stations in Amsterdam. It’s in heavy usage here in Amsterdam, so much so that if you park a car and check out of it during daytime hours your car will be used by someone else within an hour or two.

    Drivers get a reward when bringing a car to a charging station, of which there are more than enough (although even more would be better).

    With two people prices are very competitive with the city’s public transport system. It’s frequently cheaper for me and my wife to use a car2go than it is to take the tram.

  • I am aware that you are tremendously negative and pessimistic about the tech, and nothing I could say will change that. Your ego is in the pessimistic side and anything that is not would be identified by it as a treat, and you will fight over it with passion.

    Not everything in the world is chemistry. A leaf, a hair or feather are done with materials with densities similar to water, but physics(the shape of this materials, and how they are applied, like wax in the surface)make a world of difference in them.

    Today we know how to make lithium batteries with special electrodes with special treatments that store 10 times more energy in the laboratory. We know how to make ultracaps with grapheme that are 1 atom thick. This is not science fiction, this is reality, it is possible. We need to make production of those, make it economical and solve all the new problems that new tech creates, this means they won’t be in the market tomorrow but it will happen.

    As they say, people overestimate the changes of tech in the short term, then they get disappointed and underestimate the impact at medium term.

    Then you have fuel cells, which are batteries itself, and they could do the same thing you do burning things(extract energy from a redox reaction) but without Carnot thermal engine limits(80% work efficiencies instead of 40%).

    In fact you could redox long chain hydrocarbons on a fuel cell, I had done it. The only problem today is that you clog the membranes over time. The animals extract energy from redox cycles their selves without cylinders explosions.

    You could buy butane fuel cells for your laptop today. Price has to go down for using it on your car(lots of cars in France and Germany could use butane-propane), but it will happen.

By Tom Lee