CleanTechnica is the #1 cleantech-focused
website
 in the world. Subscribe today!


Batteries hawaii maui wind project

Published on January 31st, 2014 | by Zachary Shahan

108

Energy Storage Market Set To Explode

Share on Google+Share on RedditShare on StumbleUponTweet about this on TwitterShare on LinkedInShare on FacebookPin on PinterestDigg thisShare on TumblrBuffer this pageEmail this to someone

January 31st, 2014 by Zachary Shahan
 
hawaii maui wind projectEnergy storage is often heralded as the “holy grail” of the energy market. It seems that a number of researchers and companies have worked hard and long enough that this holy grail is ready to see the light. According to market research firm IHS, the energy storage market is set to “explode” to an annual installation size of 6 gigawatts (GW) in 2017 and over 40 GW by 2022 — from an initial base of only 0.34 GW installed in 2012 and 2013.

The IHS report pits the US as the largest market for grid-connected energy storage installations through 2017. It projects that the US will install 43% of the capacity additions from 2012–2017. Germany and Japan are projected to be other top markets, as any regular reader, long-time of CleanTechnica would surely assume.

What will rule the day in the energy storage market in the coming few years? Zinc-air batteries? Zinc-iron redox flow batteries? Liquid-metal batteries? Not according to IHS. IHS projects that 64% of energy storage installations will come from lithium-ion batteries. That’s more or less what a recent panel of battery experts told me in Abu Dhabi as well — a story for a coming day.

In the longer term, it’s much harder to predict what will rule the day, or how much growth we’ll see. But, for now, I’ll let Sam Wilkinson, solar research manager at IHS, have the last word:

“The grid-connected energy storage market is set to explode, reaching a total of over 40 GW of installations by 2022.”

Keep up to date with all the hottest cleantech news by subscribing to our (free) cleantech newsletter, or keep an eye on sector-specific news by getting our (also free) solar energy newsletter, electric vehicle newsletter, or wind energy newsletter.

Print Friendly

Share on Google+Share on RedditShare on StumbleUponTweet about this on TwitterShare on LinkedInShare on FacebookPin on PinterestDigg thisShare on TumblrBuffer this pageEmail this to someone

Tags: , , , , , ,


About the Author

spends most of his time here on CleanTechnica as the director/chief editor. Otherwise, he's probably enthusiastically fulfilling his duties as the director/editor of Solar Love, EV Obsession, Planetsave, or Bikocity. Zach is recognized globally as a solar energy, electric car, and wind energy expert. If you would like him to speak at a related conference or event, connect with him via social media. You can connect with Zach on any popular social networking site you like. Links to all of his main social media profiles are on ZacharyShahan.com.



  • Harald

    Has anyone looked at the external power bank batteries on ebay with a claimed capacity of 80 000mAh? If this is correct the price is around 120$ pr kWh…

  • Rick Kargaard

    I am concerned when people use the word unlimited. There is no earth based resource that is not finite. However, extraction from seawater would yield other useful by-products. Land deposits usually contain larger concentrations of other useful minerals such as potash. The lithium itself is usually a secondary reason for mining.They are, after all, ancient seabed deposits.

    • Bob_Wallace

      Certainly things are finite. But the finite point for some things is so far out there that for practical reasons we can treat them as infinite. Let me post something that I worked for EVs and lithium….

      The 100 mile Nissan Leaf uses 4kg of lithium in its batteries. Let’s say magic happens and between 2015 and 2035 we put 1.2 billion 200 mile range EVs on the world’s roads, each using 8kg of lithium in their batteries. (And that’s if range increase comes only from more batteries rather than the more likely improved anodes and cathodes.)

      That would mean that in that 20 year period we would need to produce 480,000 metric tons of lithium per year.

      And after that we could just recycle what we’ve already extracted. Even if we didn’t recycle we’d be fine.

      At 20 mg lithium per kg of Earth’s crust, lithium is the 25th most abundant element. Nickel and lead have about the same abundance. There are approximately 39 million tonnes of accessible lithium in the Earth’s crust. An 81 year supply.

      Argentina, Australia, Bolivia, Brazil, Canada, China, Portugal and Zimbabwe have roughly 13,000,000 metric tons of lithium that can be extracted. That’s a 27 year supply.

      Bolivia has 5.4 million of the 13 million tons. Over 11 years.

      There are approximately 230,000,000,000 tons of lithium in seawater. A 479,167 year supply.

      http://en.wikipedia.org/wiki/Lithium#Terrestrial

      • Rick

        so no one is mixing Lithium with anything else like cobalt and manganese? I saw earlier that the cost of extracting Lithium from seawater is a lot higher than Lithium thats been mined.

        • Bob_Wallace

          The topic was lithium. The cost of lithium wasn’t part of the topic.

          Yes, extracting lithium from seawater is more expensive. But there’s not that much lithium in EV batteries to make the batteries unaffordable.

          The LEAF uses 4 kg. Currently battery grade lithium carbonate is around $12/kg. The price has spiked up from about $8/kg which is more likely the normal price.

          4 kg at $8 is $32. 4 kg at $12 is $48. A car with twice as many batteries would use $64 to $96 worth of lithium.

          Extracting from seawater runs 3x to 5x from what I’ve read. That would make the lithium cost for a 200 mile range EV somewhere in the $192 to $480 range.

          It’s small hundreds, not thousands of dollars more for EVs 20, 30 or more years from now if we’ve used up the readily available land resources. It’s still going to be a small fraction of driving with petroleum.

          • RickSolar

            Thanks Bob for the comment, love your work. I note a lot of thinking around lithium extracted from seawater is theoretical-no ones actually done it yet. It doesnt seem to have bothered the Chinese-as they do-they are trying to dominate the Lithium supply chain by building ‘relationships’ in Chile and Bolivia, and by heavily subsidising their own electric cat production. All of this hasnt bothered Tesla and panasonic, theyve stuck with the tried and tested Lithium Cobalt. Would this have something to do with the Model S’s superior range and quick charging time?

  • Ross_UAE

    None of the technologies mentioned in this article are as cost effective or practical as pumped storage, that’s why there is up to 127GW of this proven technology installed worldwide. Modern pumped storage plants can have an efficiency of over 80% and a low maintenance operating lifetime measured in decades if not centuries. Pumped storage pumpturbogenerators are also the most effective means currently available of stabilising power grids containing a high percentage of non-synchronous generation such as wind and solar. In contrast battery storage is expensive, has a short life, and because it provides no synchronous inertia, battery storage inhibits a grid’s capability to facilitate capricious renewable generation.

    • Tinker_Noseyparker_Soldier_Spy

      I am assuming that a pump storage generator involves dam(s) and wind turbines with the wind turbines being used to pump the water back behind the dam when capacity or wind is spare a bit like those proposed to the previous Irish government for locations on the west coast but subsequently rejected?

      • Ross_UAE

        Pumped storage can use electricity generated by any means. This can be excess wind generated electricity or electricity from a conventional power plant running on minimum load at a time when the demand is very low. Pumped storage is a very valuable in Ireland as it balances capricous wind generation, more is needed as wind generation increases.

        • Tinker_Noseyparker_Soldier_Spy

          So basically, you use spare electricity to fill up a dam to meet peak demand instead of dumping the unused electricity..?

          • Ross_UAE

            Yes, that’s how it works! for each 1kW deposited at a time when additional electricity is cheap or even worthless, about 800W can be withdrawn at a time when electricity is valuable. Equally important in wind turbine obsessed Ireland is the grid frequency & voltage stabilisation benefit; when it comes to the provision of system services EirGrid values a MW of pumped storage at five times more valuable than a MW of any other power generation technology. Turlough Hill pumped storage plant was out of operation over Christmas, the result was a lot of curtailed (dumped) wind power. For the full story on the need for system services that will cost the consumer at least €360m per year (because of wind power) and the value of pumped storage go to EirGrid’s website and read the DS3 publications. The cost will be kept to €360m annually only when the power quality is allowed to drop, should we retain current standards expect the cost to be €600m p.a. The Irish electricity market is wort about €3.5b, so we are looking at an additional 10-17% on our bills

      • Bob_Wallace

        Any generator can provide the electricity used to pump water up hill. The US, Japan and other countries built a lot of pump-up hydro in order to use nuclear on the grid. It was necessary to store unneeded late night electricity since the reactors couldn’t be shut down during off-peak hours.
        Future grids might use both wind and solar to fill reservoirs when supply exceeds demand and then use that stored water to produce electricity when needed.

    • Bob_Wallace

      Agreed. But when looking at all the emphasis being put on battery storage I get the feeling that people in the industry expect battery storage to be competitive with pump-up.

      Battery storage has the advantages of being much easier to site, much quicker to install, and the ability to be distributed around the grid. Disbursed storage means more reliable local grids and less transmission issues.

  • Rick

    I know this has been discussed, and theres a great source on information on battery universe. But isnt CLMOS the best option? There is after all over 7 billion tonnes of cobalt rich manganese nodules lying on the sea in Cook Islands waters.

  • Jouni Valkonen

    I also think that lithium batteries are the mid term winners (by 2025). After that we of course cannot predict what new battery technologies are developed.

    • brink

      Once utilities understand storage economics Lithium ion will not dominate. I can tell you for a fact few of these utility personnel have a clue about where the benefits or value of the different technologies are most of them think they can use lithium ion batteries to meet their peak load period of 4+ hours without degradation or causing a fire while meeting projected life expectency

      • Bob_Wallace

        It’s far too early in the game to declare what the storage solution(s) will be. Lithium batteries may improve greatly (there’s interesting stuff in the labs) or we may utilize another technology.

        Fires, not the issue you seem to be claiming.

        • brink

          sure they are Bob check UL report. my point being you cannot have millions of batteries in that type of application and the risk of fire is not there not only that they wont last 4* hours they are not built for that but someone will sell them for that (greed)

          • Bob_Wallace

            At one time we had problems with lithium batteries and laptops. Those issues were resolved.

            New technologies often have problems during their shake out.

            At this point we’re not really addressing long term storage or large scale supply shifting. Right now the issue is grid smoothing/frequency regulation. Getting batteries into the role performed by spinning reserve. Bulk storage won’t be needed for a number of years.

          • brink

            Bob, let me educate you, most of the customers in this country are served by municipal utilities or co operatives and most of them are non generators who are paying a mortgage for assets installed years ago by their wholesale suppliers. they are paying ta mortgage in the form of peak charges (that pesky little problem) which are long term agreements (20-30) yrs. the only why to address the problem is energy conservation and efficiency which is very iffy and not consistent.

            Next best thing? bulk storage for DR , which can address peak charges every month with additional benefits of reducing line losses and performing ancillary services. We are working with a utility with a customer base of 30,000 and demand charges exceeding $225, 000 monthly, by the way the utility contributes more to the city budget than tax revenue. They are now under some budget constraints due to a lack of council actions over the years. We can eliminate that demand bill for them through our services, and contribute some profit to the city coffers. This is all over america today!!!!

          • Bob_Wallace

            You’re talking about storage at the user side of the meter.

            I was talking about storage on the utility side.

            Yes, with TOU metering storage will start to come into play on the user side. People with rooftop solar are likely to start installing storage if they get hit with high late afternoon peak charges.

            But as far as the grid goes, we’re years from needing lots of new bulk storage. The US grid can be at least 30% wind and solar as the grid now exists. We’re just now getting to 5%.

          • brink

            Bob, I used to think you knew what you were talking about but at the same timer it is unfair for me to describe that scenario to you most utility engineers have no clue unless they work for a non generator the situation I described to you IS on the utility side. If you do not know the relationship between a wholesale power supplier and the utility you should not comment or go educate yourself first. I say this not in jest or ridicule but you are ignorant on this subject based on your comment that this was a behind the meter issue

          • brink

            the utility is the customer so, yes it is on the utility side

          • Bob_Wallace

            In my case I was clearly talking about the two sides to the meter.

            Retail/industrial/commercial purchasers on one side of the meter, utilities and their suppliers on the other.

            You can easily make my statements incorrect by redefining my words to mean something different than I intended.

            You seem to have some sort of bug up your ass that is driving you to misinterpret me.

          • brink

            no bug and no need for profanity. you cannot claim i made a mistake when you joined the thread maybe you should have paid more attention to detail since we are talking about grid scale storage. just admit you have no knowledge in this regard and call it a day

          • brink

            even in this case i am talking about the utility side of the meter so i did not misinterpreted you, you just do not have a clue. Bulk storage is needed now because billions of dollars are being wasted. By the way it will give relief commercial behind the meter as well, duh!!!!!

          • Bob_Wallace

            You clearly have a writing problem.

            Do you also have a reading problem?

            Billions of dollars are being wasted on the utility side of the meter? We’re curtailing billions of dollars of wind or solar production?

            If that’s not the case where are these billions being wasted?

          • brink

            sure it is Bob there are billions in co incident peak demand charges that are being paid by small utilities everyday to wholesale power organizations. like I said you are ignorant on this matter

          • brink

            i won’t give you anymore information you might misuse it. Lets just say everything cost money, you want to be pretend wind and solar do not have capital costs and they just start generating power. That capital cost is what we call a mortgage and someone has to pay for it.

          • Bob_Wallace

            What is the matter with you?

            Why would you post something as stupid as that?

          • Bob_Wallace

            I may be. But you are doing nothing to educate people, just throwing out numbers.

          • brink

            its called comprehension Bob, you took that in school, right?

          • brink

            they are not solved , there are incidents the media does not cover down load this underwriters report on li ion

            http://www.renewableenergyworld.com/rea/companies/ul-underwriters-laboratories/white-papers

            the size of these installations with millions of batteries makes it inevitable there will be a failure and not always a catastrophic one but a failure non the least. which speaks to what applications are they good for certainly not multiple hours of deep discharge 2-3 times a month we need a more versatile alternative but people seem to be caught up in the sexiness of this chemistry but the people in the know are not so enthusiastic. there are numerous former li ion executives who have moved on to other chemistry they see something better long term

          • brink

            the chemistry lacks durability and is highly volatile but not always flammable but does rupture . It lacks performance if abused and will completely die if charge is not kept. How long does your cell phone last? or battery laptop? take it off of charge for about two months and see if it has not ruptured

        • brink

          labs? really Bob much more significant breakthroughs are happening with other technologies that will have a major disruptive effect that is near term

  • Doug Cutler

    Anyone here have an opinion on pumped heat storage system from Isentropic in the UK? They claim 70-80% round trip efficiency at crazy low cost – just a couple of big silos full of gravel. Currently in scale proof of concept trials in UK. Is it just crazy talk or what? I wish I was an engineer but alas I’m just a musician but can still sorta manage entry level discussion.

    • Bob_Wallace

      They got funding to build a system and the resulting data should tell us whether their idea works or not. Many things sound good, a few work out.

  • JJ

    I hope if they try to achieve anywhere near the 64% number using Li-Ion technologies that the energy storage market doesn’t explode (article title?)…

  • anderlan

    Assuming 2 packs per station (Musk said half a MW each, so that’s 2 packs, or 6 packs if he meant half a MWh) Tesla already owns 43MW. Packs per station will probably be beefed up along with car sales, but first buildout will increase their total.

  • jeppen

    The power numbers (GW) are interesting, but even more interesting would be numbers for energy (GWh). Anyone?

    • anderlan

      That’s easy to narrow down. We know the energy and power density ranges of lithium batteries:
      http://en.wikipedia.org/wiki/Lithium-ion_battery

      Assuming a low end of energy density and–pessimistically, not optimistically, for our extrapolation–the high end of power density per kilogram (100Wh/kg and 340W/kg) the number is, at minimum, 11.8GWh.

      We could also take the high end of energy density and the low end of power density (265Wh/kg and 250W/kg) and get 42.4GWh at the top end of our envelope, assuming technology stops, which it won’t.

      • jeppen

        Hmm, ok, that’s good thinking! The world is producing some 23,000 TWh electricity/year, so 42 GWh is less than one minute of current global electricity production.

        • A Real Libertarian

          And the entire global electricity production has to be run off of storage… when exactly?

          Oh, right! When nuclear becomes a good idea.

          • jeppen

            You mean you are going to ship storage around? I don’t think that’s feasible. If you have a grid that is dominated by intermittent power, say 80%, you’d need at least a few weeks of storage to be able to handle varying supply.

          • Bob_Wallace

            Please, jeppen, don’t post your ignorance.

            You could save everyone some time and effort by simply posting “I have an unnatural love of nuclear energy” over and over.

          • jeppen

            I didn’t talk about nuclear. This is a post about storage, apparently motivated by the obvious fact that high volumes of storage is a necessary enabler for high-penetration intermittent storage. You know, the “holy grail” thing the post mentions?

            I just pointed out that this small amount of storage is kind of depressing. Optimism is generally a positive thing, but too much optimism for inadequate solutions in a disaster situation is obviously dangerous.

          • Bob_Wallace

            There is no need for “weeks” of storage.

            Nuclear needs storage and backup as well.

          • jeppen

            I remember challenging you to tell how you’d like to calculate storage needs, and I’d then simulate it using some real-world nuclear and intermittent power production data and provide the results. I think you met that with silence.

            Of course you need weeks of storage for intermittent power.

          • Bob_Wallace

            That’s just crazy. Read the Budischak paper. They used four years of hour by hour real world wind and solar data and never did they need “weeks” of storage.

            BTW, I don’t have the time to reply to the very large number of posts you make. I’m going to ignore a lot of your foolishness just out of necessity.

          • jeppen

            You have a very clear pattern of avoidance. When you call me dishonest and are proven wrong (happened several times), you don’t have time to say sorry. When your numbers are wrong, you don’t have time to admit it. When I try to pull you a bit deeper into details, you don’t have time to follow.

            But you have plenty of time for shallow posturing, blatant trolling and for labelling what others write “foolish”, “stupid” and so on. It’s quite sad to see.

          • Bob_Wallace

            I’m not going to get into a pissing contest with you.

            You’ve already demonstrated that you put a very low value on human life and that means that your opinions are worthless to those who care about others.

          • jeppen

            No, you don’t get into a pissing contest with me. You just piss on me, period.

          • Bob_Wallace

            I have a very low regard toward someone who cares so little for others.

          • jeppen

            Pathetic troll is pathetic. Especially the troll who is fine with millions of particulate deaths and the existential threat of AGW in the hope that his beloved tech will one day combine to provide baseload power.

          • brink

            Amen, he does that a lot.I for one will no longer respond to his posts

          • A Real Libertarian

            “You mean you are going to ship storage around? I don’t think that’s feasible. If you have a grid that is dominated by intermittent power, say 80%, you’d need at least a few weeks of storage to be able to handle varying supply.”

            Nukes need storage.

          • jeppen

            What does that have to do with anything? Kind of off-topic, but the obvious reply is “not of the same magnitude”.

  • Omega Centauri

    You said 64% of installations. Does that leave scope for more than 36% of capacity being other than Lithium batteries? I have the impression Li-ion, are great for small to moderate sized voltage/frequency regulation, but for load shifting some other technology is probably in order.

    • driveby

      LiFePO4 is being deployed in outback Australia for off-grid now and preferred to any other technology.. 1-2 more years and those systems will reach grid parity for a 10 year ROI.

      Next big thing for the masses will be Solar Hybrid Inverters, which incorporate an off-grid/on-grid solar converter and a battery charger/inverter in a single box. Just add batteries and you can decide how much time you will be spending off-grid.

  • rex

    Rhubarb batteries(really) the quinones are a substitute for vanadium and could potentially reduce costs by 2/3:

    http://news.sciencemag.org/chemistry/2014/01/rhubarb-battery-could-store-energy-future

    • SirSparks

      Add some custard to it for a really tasty battery.

  • jimbo

    It’s great that energy storage is coming of age. I do think the types of storage will differ depending on what function it will serve. Batteries may be more useful for short duration load leveling etc. As a UK resident with a solar roof, It is becoming apparent to me that if solar PV is adopted widely in the UK, then there would potentially be enough for a large contribution in the summer but bugger all in the winter! A possible solution could be a large production of hydrogen, or preferably synthetic methane which is made during the summer and stored [in old exhausted natural gas wells?] and this could be used during the winter in CHP gas turbines.

  • nullbull

    There are many ways to store energy that don’t involve sticking as many electrons as possible to a thing, and then pulling them off that thing later. For energy sources that begin as heat, we can store them as heat (LMB, but also molten salt, efficient boilers, etc.). Also, the most efficient storage medium depends on the output need – are you going to use the energy to heat something? Is the source of energy a thermal source? Then why convert it from heat to electrons and back to heat?

    In my mind, a huge potential market for energy storage comes from heat pumps, waste heat harvesting, and bi-directional partnerships between symbiotic installations (e.g. A heat-using facility, like a smelter, sharing with a waste heat-producing facility). These aren’t grid scale, but the nearer the closed loop to the source and destination, the more efficient. The more efficient, the more economically viable. Not sure if this kind of “energy storage” is counted here. Probably just a semantic/categorization thing.

    • driveby

      you have to factor in that electrical energy is quality wise better and higher than just heat. It’s easier to convert electric energy into heat or work or do something than it is with heat.

      So when you just need heat in the end it makes sense, yes, but for everything else it’s better to store electricity so you wind up with a more desirable, more diverse useable source of energy.

  • Steeple

    It seems like a difficult question to answer how much storage any particular geography needs. In the US, there is clearly a market signal that demonstrates that peak hours power is worth more than off peak, and therefore there is an incentive to store off peak and use during peak times.

    Once the price delta shrinks to a level where investment in batteries doesn’t pay, we’ve probably built enough storage. Combine that with the potential for material EV charging overnite and it’s difficult to determine how much of this we will need in the medium term.

    • driveby

      You’re missing the bigger picture there Steeple..

      Your point is only valid until storage cost comes down so much (or power from the grid goes up so high) in comparison to solar/wind that your price delta between off-peak and on-peak is not the hot-spot anymore and people try to use their solar/wind produced energy when the original source isn’t shining/blowing.

      Once it’s cheaper to store you locally produced energy for the next 48 hours, people will do it. It then doesn’t matter if the off-peak/on-peak price delta is there or not.

      • Steeple

        I’m just thinking that you store in times of surplus and withdraw during times of need. The economic signal of pricing these periods helps to allocate the storage resource most effectively. My thoughts are on a macro market basis and not specific to a single installation.

        • brink

          Steeple, I think the point driveby is trying to make is that storage has a number of value streams other than just shifting energy between on and off peak and he is correct that the cost to store energy (O&M) will come down with reduced battery capital expense.

          Some battery O&M is already below 8 cents way below solar and some wind so, the nexus has to do with the cost of generating solar dropping faster than the drop in the cost to store energy in batteries.

          Power quality is a major concern in american industry and thus a low cost to store energy battery will always have value great value to this segment even if the cost to store is greater that the cost to generate. Poor power quality leads to down time in factories which have a significant impact on revenues plus batteries give the added benefit of voltage conservation, back up and start up capabilities

          • Steeple

            Fair point. It could be extremely valuable in manufacturing as you mention. I think the value proposition for a residential user is likely a lot lower.

          • brink

            the back up value would apply to the home and give grid resiliency. i could see where the utility could pay you to provide demand when solar is not producing or during major storm night fall and solar is not producing or panel is ripped off. diesel generator does you no good if roads are blocked (no Fuel) so for peace of mind while performing in between it is good

          • Steeple

            Definitely would rather invest in battery backup than a generator

          • Bob_Wallace

            Residential electricity cost is a lot higher than for industrial users. Make sure you compare costs on an apple:apple basis

          • brink

            Bob, a quick comparison here (Florida) shows that is not true in most cases you have to factor in the peak demand charges the are placed upon industrial users and not residential.

          • Bob_Wallace

            Sorry, I was replying from email and thought the discussion was about German prices.

          • Bob_Wallace

            “Some battery O&M is already below 8 cents way below solar and some wind”
            Not sure what you mean here. The O&M for wind and solar is around 1 cent. The cost of wind generation (LCOE) is around 5 cents.

          • brink

            my mistake i interchanged the O&M for generation which is take into account in storage speak

          • brink

            not all wind is at 5 cents

          • Bob_Wallace

            The average selling price for onshore wind in the US in 2011 and 2012 was 4 cents per kWh. Add in about 1.3 cents for the PTC.

            That puts the average cost of new onshore wind in the US at about 5 cents.

  • Will E

    storage in electric heat pump house heating and storage in EV car batteries.
    to start with.

    • driveby

      to shallow.. there are regions which need to store cold instead of heat ;-)

  • jburt56

    Should be GWh, not GW.

  • Senlac

    I really hope the other than lithium-ion technologies pick up the pace. I just don’t think using lithium-ion for grid storage is the best use of what is a limited resource.

    • http://zacharyshahan.com/ Zachary Shahan

      you mean lithium being limited? hardly.
      however, the production of the batteries still has some clear bottlenecks.

      • Senlac

        Well if lithium isn’t limited that is good news all around. I remember hearing there was a concern. Perhaps more has been found.

        • http://zacharyshahan.com/ Zachary Shahan

      • Cwacks

        Perhaps no problems just yet – but in the near future there may be:
        http://www.meridian-int-res.com/Projects/Lithium_Microscope.pdf

        • Jouni Valkonen

          There are limitless amount of lithium in sea water. And it just costs about 3–4 times more to distill lithium from sea water than to mine it. As lithium contains only small fraction of total cost of battery, this price difference is negligible.

          And moreover, it is very likely that the cost of extracting lithium from see water will plummet, when surplus production of renewables pushes the electricity price time.

          • Cwack

            Read the report.

          • Jouni Valkonen

            That is not a report, but it is utter polemic nonsense. It even did not try to make arguments based on facts, but it is just handwaving and tries to make a “convincing” point with absurd and snarky analogues.

            Of course production lithium from seawater is expensive. It costs as much as three times more as the production of lithium from geological deposits.

          • Bob_Wallace

            The Nissan LEAF uses 4 kg of lithium. Scale that up to a ~200 mile range EV and it’s somewhere around 12 kg.

            A few months back battery grade lithium carbonate was selling for $8/kg. Now it’s up to $12. So $96 to $144 per car for lithium.

            Extract from seawater. Now we’re in the $300 to $500 range. That is not exactly a deal breaker when you’re saving around $100/month by not having to purchase fuel.

          • Sachas Perez

            I have a question: If we pay for tonn litihium carbonate 6000 $, and we use use this material in the battery around 10 kg of lithium carbonate, is correct to say that the cost of this litihum carbonate used in the battery is only 60 $. I think that’s a retail value of lithium compounds, but without a conditioning particle size used in a battery. So this extra step means a higher cost. Is this correct or not?

          • RickSolar

            Why will it plummet? It doesnt with desalination. Has anyone put the theory of extracting Lithium from seawater into practice?

          • Jouni Valkonen

            Koreans have invested about $30 million on pilot plant. It should be operational around 2014. So the technology is well studied and there are several methods how to do it, but as I said, it is not yet commercially viable due to too high cost of electricity.

            South Korea commercialization Lithium-Extraction from Seawater
            http://nextbigfuture.com/2011/01/south-korea-commercialization-lithium.html

          • RickSolar

            haha how ironic. This is the same reason why economies of scale cant be achieved with desalination. Its simply cheaper to obtain bottled water. I believe the Koreans have theorised a plant, like the studies. And it is now ‘around’ 2014. The answer to relying on high concentrations of Lithium is to mix it with something else more abundant like cobalt and manganese-of which there is 10 billion tonnes of nodules lying on the seabed in The Cook Islands.

      • RickSolar

        Why is it everyone brazenly suggests that Lithium is not a limited resource? Are you assuming extracting it from seawater will work? GM, Toyota and Tesla have gone with Cobalt Manganese do they know something you dont?

        • Bob_Wallace

          Yes, after 20, 30 years of building EVs with lithium from surface sources we’d have the option to use lithium from sea water. Or recycle it from used batteries.

          The sea water lithium would be 3x to 5x more expensive than lithium salts sourced. But it would still not be very expensive. The Nissan LEAF has only $24 to $48 worth of lithium in its batteries.

          (The price has fluctuated between $6 and $12 over the last few months. Demand seems to be getting ahead of processing. Processing will catch up and the price drop back down)

          • RickSolar

            Hi Bob, Im not suggesting using Lithium is a bad choice or the price/economics of it, Im basically saying by mixing it with cobalt and manganese as a composition it achieves two things 1. Makes the existing Lithium resource last longer. 2. Encourages the use of more diverse resources such as Cobalt rich manganese nodules found on the seabed in places like the Cook Islands. This also diversifies the threat that these rare minerals are dominated by China, and are extracted from dodgy places like the Democratic Republic of Congo. It shares the growth and success of the rise of electric cars and smart grids with other countries like the Cook islands. All this is aside from the fact that having Cobalt rich Manganese makes for a better electric car battery.

          • Bob_Wallace

            And I’m not making any comment on cobalt and manganese. I have no clue as to what will make the best EV or grid storage batteries. (I sort of think lithium may not be the grid solution, but that’s nothing but a hunch.)
            There are so many things being worked on in the labs. Who knows what we’ll be using 3, 5 10, 20 years from now?

            All I’m doing is pointing out that if we go with lithium there should be no shortage.

          • Otis11

            I can actually really see Li-ion being the main distributed storage… Run it in cars for 7 years, then spend 15 years on the grid before it’s recycled. My biggest argument for something else is because we won’t have enough used batteries to keep up with demand…

        • http://zacharyshahan.com/ Zachary Shahan

    • brink

      Senlac, you are correct but not in the way you think, Litium ion batteries do not capture all of the value streams available to make a project successful at current pricing. Capturing multi value streams will accelerate deployment, yet supply constraint is a concern with the newer technologies looking to roll out.

      Lithium ion batteries are used mostly for power applications (frequency regulation and voltage control) but will have major difficulty meeting energy services demands of 4+ of duration deep discharge for Demand Response(DR) 3-4 times a month while shallow cycling in between. That type of activity will greatly reduce the life of the battery and accelerate replacement before projected time.

      • Ross

        Isn’t this just a bit of battery management software to ensure that it isn’t the same cell being cycled all the time.

        • brink

          Ross,
          No that is not the case when performing demand response you need all of the battery capacity to meet service requirements and contract terms. As I stated earlier cycling Lithium ion batteries starts the degradation and aging process by which capacity losses are realized and safety issues become a significant concern

          • Ross

            Got anything to back up that claim? Deep discharge of a large number of cells or cycling through a small number of individual cells at a time in the battery shouldn’t be mutually exclusive modes of operation for Lithium-ion batteries.

          • brink

            Ross, a car battery is not a storage contract with required performance terms. You cannot use Li ion on a 20 yr capacity contract they degrade, it is a fact that there are annual capacity losses in all types of li ion batteries to what degree it varies. It is also a known fact that heavy usage (cycling) and deep discharge accelerate that process. In a demand response application you may have to displace lets say, 5,000 kW which is 5(MW) which is the minimum performance because projected annual customer is 1.5%.Lets say battery degradation (capacity loss) is 10% in 3yrs even with no customer growth you have breached the contract minimum performance requirements. I know what you are going to say well we will just add more cells well you could but now you have a very expensive battery system with high maintenance cost due to annual down time and a risk of failure.Economically not feasible

            http://www.renewableenergyworld.com/rea/companies/ul-underwriters-laboratories/white-papers

            go to above page and read UL report while risk of fire occurs every so many millions that is for high quality manufactured batteries and keep in mind the size of these installations with many millions of batteries and the effect of aging and usage accelerating that process. No you are not going to be moving through a small window of cells at a time where’d you get that from?

          • brink

            I was referring to the mostly one dimensional use for li ion that a utility purchasing that system does not understand where all the value streams in storage are. If you are a transmission operator that is a different scenario but if you are a generator then I would argue you need a more diverse piece of equipment to realize maximum revenue at the distribution level if you are a non generator it is required that you use an alternative to li ion for capacity and to serve your biggest customers

      • Senlac

        A good point, we need more battery technologies which can best address the unique challenges of grid storage. In addition we need more capacity in general in the pipe line, that can eventually come from a different technologies once they ramp up.

        • brink

          Senlac, glad to know someone is knowledgeable about storage economics on this blog, look forward to spirited debates and conversations.

Back to Top ↑