Blog Post

Vinod Khosla On Why Lithium-ion Batteries Are Overhyped

khosla-alwaysonVinod Khosla, co-founder of Sun Microsystems (s JAVA) and one of Silicon Valley’s most active cleantech investors through his firm Khosla Ventures, doesn’t always sound like a VC who has invested hefty sums in lithium-ion battery startups, including Sakti3 and Seeo. “Lithium-ion batteries are overhyped and will possibly be replaced,” he said today at the AlwaysOn Summit at Stanford. He told me after his talk, however, that Khosla Ventures is backing the technology because the “lithium-ion markets are here today. We’re investing because there are good markets.”

Problem is, costs have to down — or oil prices have to shoot up — for most battery ventures to make big winners of entrepreneurs and their investors, Khosla said. When asked during his talk this morning with Fenwick & West Chairman Gordon Davidson what it will take to get low-carbon cars on the road in significant numbers (which would boost demand for technology from companies like Sakti3), Khosla said, “The most important thing to remember is economic gravity — the cheapest thing ends up winning.” For that reason, he said, “I can’t imagine lithium ion making low-carbon cars pervasive anytime soon.” Then again, if oil prices jump to $100-150 per barrel, “you’ve got a different ballgame.”

It’s largely the cost issue that Khosla says makes hybrid vehicles “greenwash.” He drives a hybrid, and considers the Toyota Prius “interesting,” but says it offers a very expensive, and thus limited solution for reducing greenhouse gas emissions. “Just get a bucket of white paint, paint your roof, and you’ll save more carbon,” he said, than you would driving a Prius. Never mind all-electric cars, he said, which in the U.S., China and India will be “plugging into a lump of coal” for years to come.

Khosla sees government subsidies in the U.S. and Europe as a key to bringing many clean technologies to the global market at competitive cost. “Every investment we make should reach unsubsidized market competitiveness,” he said, usually in the timeline of 5-7 years. This is important, he said, because fast-growing markets in China, India and elsewhere in the world don’t have subsidies. So the idea is to “use the U.S. and European regulations to get down the cost curve until such time that you can compete” without government assistance.

At this point, Khosla sees some parallels between the relatively early stage of cleantech innovation, and expectations for the Internet in the early 1990s — we have no idea what “one breakthrough will change the rules on what renewable energy is.” Back in 1990, he said, “everybody assumed the digital world would be interactive TV…the Internet came from left field, from an unlikely instigator: the browser.” Right now, it’s too early to tell what the instigator will be for energy. “There are radical approaches that might come along, a different kind of battery, for example…or a different kind of sequestration, that we aren’t planning on.”

On the road to those radical breakthroughs, we can expect to see some massive failures in the cleantech sector. “What is amazing about this is the size of the markets. We are dealing with much harder science and technology, so we will see much higher rate of failure,” Khosla said, “but the wins will be bigger. More money will be made in cleantech than in traditional areas of Silicon Valley — by far.”

32 Responses to “Vinod Khosla On Why Lithium-ion Batteries Are Overhyped”

  1. Henry Gibson

    The ZEBRA battery can outperform many if not most lithium batteries for automobiles. Lead batteries have been proven to be adequate for plug in hybrids, and this was before the Firefly battery. EFFPOWER has a lead battery for high power for straight hybrids. ..HG..

  2. owlafaye

    Clean electricity (nuclear, wind, solar, hydro) and dirty electricity (coal) produce, lets say, 100 units of electricity each year. Using clean electricity to charge your car simply means that that unit you would have used from a dirty source is being used by another consumer…there are still 100 units being used a year.

    I like all the foolishness surrounding this…pure hilarity. Keep it up: Laughing is good for the planet.

  3. @Josie Garthwaite

    Re: Your comments about plugging in to a dirty grid.

    Put these facts at the top of every post about EVs.

    1)ICEs are 15% energy efficient at the wheels… that’s 85% of all fuel sold being converted into waste heat.

    2)The dirtiest coal power station in the world is 40% energy efficient. (and less than 50% of US generation is coal)

    3) An EV is 80% energy efficient just driving down the road. Add aggressive regen which can re-cycle up to 50% of the energy used to accelerate and an EV approaches 100% efficiency (it’s actually 100%+ but that statement causes endless debate)

    A kw/hr of energy costs approx the same be it from a fuel bowser of a wall socket, yet it costs $0.01 per mile to power an EV while an ICE costs 20x as much. i.e the same energy used to drive 1 gasoline powered car can power up to 20 EVs doing the same work.

    The ICE is the automotive equivalent of an incandescent light bulb in terms of energy efficiency.

  4. Lithium is actually common. There is no indication we cannot support an EV revolution based on Li batteries. The concept of “peak Lithium” is more www rumor than fact. Lithium is so abundant it is usually produced from concentrated sources in salt flats and mined mineral sources are not exploited. (search “lithium Keith Evans” for more info) As a result of lithium’s abundance, Li based batteries will continue to drop in price, just as cell phones and home computers did.
    What Mr. Khosla is correct that something even better might come to the market and change Lithium’s current dominance. That may be a reason to call it over-hyped. The current price is not. Mr. Khosla is old enough to know calling Lithium batteries too expensive at this stage is the same as the IBM executives who claimed the home computer was too expensive to catch on. IMO he is wrong.
    Another point: Lithium is recyclable into new Lithium Ion batteries. i.e. It does not get used up.
    Still another point: Mr. Khosla does not mention Range-Extended Electric Vehicles (REEVs), which use smaller batteries to achieve more than 90% of the benefit of full Battery Electric Vehicles (BEVs). This will be a major factor in the insipient EV revolution.

  5. Golphnut

    One element not talked about that is considered better than Lithium is Vanadium. There is an abundence of this and has been proven to extend use in batteries and provide more power. Vanadium can be used to
    strengthen steel, store more energy in batteries and charge the battery during use. Vanadium can also be used with Lithium.

    Vanadium is the answer to all of the questions I suggest everyone to do research on Vanadium to see for yourselves what incredible potential.

  6. Dag Johansen

    Vinod Khosla has the general idea correct . . . that Li-Ion battery prices must come down or oil prices must shoot up. But he’s got a few things wrong which make things less pessimistic for Li-Ion batteries than he indicates:
    1) It is not ‘or’ it is and/or . . . it will be a combination of li-ion battery prices come down (a little bit) and oil prices going up (to around $100/barrel) that will make li-ions useful.
    2) “if oil prices jump to $100-150 per barrel, “you’ve got a different ballgame.”” is correct . . . and will happen eventually. $100+ oil may happen relatively soon if the global economy recovers.
    3) “Never mind all-electric cars, he said, which in the U.S., China and India will be “plugging into a lump of coal” for years to come.” is misleading. It is well known that due to the efficiency of electricity creation, distribution, and EV efficiency, driving a 100% coal-powered EV generates less carbon than gas car which emits a lot of carbon during crude oil distribution, crude oil refining, gasoline distribution, and gasoline usage.
    4) “because fast-growing markets in China, India and elsewhere in the world don’t have subsidies.” is totally incorrect. Both India and China do have EV subsidies.

    So I think the future for li-ions is much better than Khosla thinks. HOWEVER, I think he is wise not to invest in little li-ion startups in the USA. I think China/Taiwan/Japan/Korea will dominate this market due to first mover advantage, mass production, more experience, etc.

  7. Josie Garthwaite

    @drivin98 The U.S. does have “a lot.” We also use and import a whole lot. The last figures I saw from the U.S. Geological Survey said the U.S. has 760,000 tons of the world’s 13.8 million tons of “identified lithium resources” (known quantity, quality and grade). I’d be interested to know about other data you think shows a more complete picture. But yeah, we’re not on track to run out of lithium any time soon.

  8. drivin98

    That’s not actually a very accurate assessment of the lithium situation. We have a lot of lithium deposits right here in the US. Mitsubishi is planning on sourcing its lithium from Canada. Here’s a couple URL’s to check out

    The plugging into coal thing is more of a meme than actual fact as well seeing that if millions of cars, plugged in at night, wouldn’t sigificantly add to the amount of coal burned.

  9. Josie Garthwaite

    @Alexander Ainslie – The fact that lithium is a limited resource, with more than half of the world’s deposits being in Bolivia and most of the U.S. supply now coming from Chile, opens a slew of interesting questions about batteries for plug-in cars (I looked at some of them for this post: ). In the spirit of innovation, however, there’s also opportunity: battery makers whose technologies use less of the mineral could enjoy an advantage.

  10. I agree with Vinod that lithium-ion power is overhyped.

    One has to look at this ecosystem from the bottom up (identifying cheap, abundant, sustainable and accessable material resources) and then apply the appropriate technological solution to take it downstream.

    The source material for batteries is key to keeping the cost of future electric and hybrid cars down to affordable levels for mass market adoption.

    The unsexy facts of Lithium-Ion Power:
    The cost of Lithium has gone from $300 to $3000 per ton and, at current production levels, we will run out of supply of this resource by around 2015. By then the cost of Lithium will be so high that the cars produced using Lithium Ion batteries will be too costly for the average consumer – an important point made by Vinod. This model doesn’t scale to have any meaningful impact on the climate crisis we have created for ourselves.

    The geo-political facts of Lithium-Ion Power:
    The principal country that can fix this shortfall in Lithium resources is Bolivia who have the world’s largest untapped sources for the Lithium metal under their salt flats.

    The Bolivians, however, have been raped for so long by foreign mining companies that they are now taking the rightful stand that they do not want to allow the same thing to happen to their land and peoples all over again in relation to “developing” and not “exploiting” their Lithium riches.

    It is an interesting fact that the French corporate raider and industrialist, Vincent Bollore (see: & ), is at present all over the Bolivians promising them the moon in exchange for the rights to “exploit” their Lithium resource for the benefit of his industrial battery/supercapacitor venture, Batscap (he has also acquired the Canadian firm Avestor adding further production capacity in North America in 2007!), which will also benefit his nascent European electric car business, BlueCar, in partnership with design/engineering powerhouse, Pininfarina (see: ( & ( & ( ).

    So the US is going to have to cozy up to the Bolivians right quick if they intend to use Lithium as a source or identify another source or material. Good luck with that as the US is not very popular in these countries. Additionally, Bollore’s deal making talents in developing countries may make it hard for the US to match or beat the terms he is surely offering the Bolivians.

    In the short term the most carbon efficient car you can buy today is a used car. New cars, electric or hybrid, use up a lot of energy to manufacture and therefore produce a much bigger factual carbon footprint than an existing used car (let’s call them pre-owned, sounds more palatable!).

  11. I second @Tyler Hamilton’s comments about electric powered cars being greener even if they are powered by coal power station. Some electric cars can quite easily consume 5 times less energy than combustion equivalents. Petrol for sure is not 5 times cleaner than coal!…

    Furthermore, the cost angle is an interesting one. For sure, price is probably the main driver of the market, but significant cost savings for electric vehicles already exist. A REVA G-Whiz Electric vehicle costs around $0.02 a mile, compared to around $0.15 for a more standard American vehicle. This corresponds to an annual saving for the average american of around $1,500 (The average american drives approximately 12,000 miles a year).

    Also, with these cost savings it is possible to purchase electricity from a supplier that uses renewable energy and still have much lower running costs and a virtually zero carbon footprint.

    For more details, please consult this blog post I wrote:

    Thank you,

    Brendan @ PlentyWays

  12. Josie Garthwaite

    @Tyler Hamilton – He mentioned ultracaps as one technology that might end up being viable, and yeah, he’s still keen on biofuels. You’re right that even coal-fired power plants can be a step in the right direction compared with gas (one stat I’ve seen from MicKinsey is that taking a gas-powered car off the road in China and replacing it with an electric one would cut the vehicle’s greenhouse gas emissions by 19 percent). So you make some really good points. In general though, I think we can use the reminder that plugging cars into a dirty grid doesn’t offer a complete solution.

  13. Would be nice if Khosla talks about what’s better than lithium-ion powered cars. Is he still arguing biofuels, or does he see ultracapacitors or some other battery chemistries gaining dominance. His comment about “plugging into a lump of coal” is just ridiculous, given that every major study I’ve seen says that even if your electric car was charged 100 per cent with coal power it would still be cleaner than burning gasoline. But the fact is, it isn’t 100 per cent coal power, and some jurisdictions are much cleaner than others. I mean, Ontario’s electricity is already about 75 per cent emission free, and overnight it’s well over 95 per cent (hydro and nuclear), so electric cars win hands down.

    • By the time there are significant numbers of plugins on the road there could be large numbers of the hybrids running on E85(85 % ethanol) which reduce the co2 on the ICE side of the equation significantly.
      Electricity from the coal plant would experience losses before reaching the car motor (including grid distribution losses, losses in charging/discharging of the battery, etc.). I’ve seen fast battery chargers with less than 50% efficientcy.
      Also the plug-in hybrid must drag around the weight of the larger batteries, which the normal hybrid does not require.