#1 cleantech news, reviews, & analysis site in the world. Subscribe today. The future is now.


Published on March 4th, 2015 | by Guest Contributor


Energy Storage Could Reach Big Breakthrough Price Within 5 Years

March 4th, 2015 by  

Editor’s Note: The Deutsche Bank report is not public, but was shared privately with the author.

Originally published on RenewEconomy.
By Sophie Vorrath

A major new Deutsche Bank report has predicted that energy storage – the “missing link of solar adoption” – will be cheap enough – and technologically ready – to be deployed on a large-scale within the next five years.

The solar industry report, published on Friday, said that while costs for the greater majority of available battery technologies remained prohibitive, economically competitive batteries were the “killer app” and the “holy grail” of solar penetration.

But with many costs already lower than published literature would suggest, Deutsche Bank believes this ultimate solar and renewable energy goal might not be far out of reach.

Screen Shot 2015-03-03 at 11.14.23 AM

“Using conservative assumptions and no incentives, our model indicates that the incremental cost of storage will decrease from ~14c/kWh today to ~2c/kWh within the next five years,” the report says.

“When overall system cost decreases are considered, we believe solar + batteries will be a clear financial choice in mature solar markets in the future.”

deutsche solar storage

Currently, according to Deutsche, the cost of a typical lead-acid battery may be as low as ~$200/kWh, while best in class lithium-ion technology was producing commercial/utility packages in the ~$500/kWh range at end 2014 – half the cost of the ~$1000/kWh 12 months prior.

“We believe 20-30 per cent yearly cost reduction is likely (for lithium-ion batteries), which could bring (them) at commercial/utility scale to the point of mass adoption potential before 2020,” the report says.

Screen Shot 2015-03-03 at 11.14.52 AM

Deutsche points to the commercial-scale market as one of the first areas where battery deployment will flourish, due to clear economic rationale.

“Commercial customers are often subject to demand based charges, which can account for as much as half of the electric bill in some months,” Deutsche says.

“We think companies with differentiated battery solutions coupled with intelligent software and predictive analytics that work with the grid to avoid these charges and smooth electric demand will pave the way for mass adoption.”

The report also points to utilities as a major market for batteries on a large scale, as costs drop and distributed renewable energy generation deployments increase.

On the residential level, the report said households were still unlikely to go down the energy storage path in the short term, without proper pricing mechanisms in place, or access to solar plus storage energy packages.

But again, Deutsche sees this as as a major, untapped opportunity for utilities: “Over the next decade, we see a substantial opportunity for utilities to utilize smart grids through residential battery aggregation.”

Properly incentivised, the report says, utilities could begin to aggregate neighborhoods of solar + batteries to behave as a single source of load reduction.

Screen Shot 2015-03-03 at 11.21.50 AM
Screen Shot 2015-03-03 at 11.22.12 AM

“Batteries could be dispatched as needed to reduce peak demand across the system. In a high grid-
penetration scenario, this could reasonably lower the necessary capacity from conventional generation sources.

“In turn, we think it is reasonable to hypothesize that lowered capacity needs from lowered peak demand would simultaneously lower the need for large up front capital investment in peaker plants.”

Deutsche sites two likely scenarios that would enable this sort of utility-driven household battery deployment: Third party leasing companies and individuals work with the utilities; or a shift in regulatory framework that allows utilities to include residential solar in their rate base.

“Both of these scenarios would likely significantly improve reliability, enable microgrids to function as needed, and improve grid resiliency during emergency situations,” the report says.

So what sort of batteries will homes, businesses and utilities be using? As the table below shows, there is quite a range of key technologies that, according to Deutsche, have the potential to be longer-term energy storage solutions, subject to technological feasability and cost.

Screen Shot 2015-03-03 at 11.14.19 AM

And while lithium-ion and other electrochemical based batteries are still the most commonly discussed – AES Energy calls lithium-ion as the chemistry of choice for the next decade – the Deutsche report details a wide range of potential new technologies looking to fill the need for on grid storage.

Screen Shot 2015-03-03 at 11.20.53 AM

Flow batteries, for example, while a relatively new technology and probably not likely to make a big impact on the market for several years, are tipped by Deutsche to be suited to large-scale utility storage with potential for long-term adoption.

Typically consisting of two tanks of liquids (electrolyte) which are pumped past a membrane held between two electrodes to store and generate electricity, flow batteries have the advantages of ease of scaling, reliability, and long life.

Varieties of flow batteries include Iron-Chromium, Vanadium Redox and Zinc-Bromine. Deutsche notes that recently, EnerVault dedicated its first commercial flow battery-based energy storage system in California.

In Australia, Brisbane-based company Redflow is fast-tracking the rollout of the latest iteration of its unique zinc bromide flow battery to the residential and mining sectors, the costs for which, it says, are 40 per cent cheaper than its first generation products, and are now approaching grid tariffs in some markets.

Elsewhere, start-ups are raising funds to scale up manufacturing of products such as EOS energy’s hybrid Zinc cathode, aqueous electrolyte based battery.

Once fully ramped, EOS says the battery will have 75 per cent round trip efficiency, a 30 year lifetime, and a cost of $160/kWh. As Deutsche points out, zinc is a much cheaper material than lithium, but has problems with electrode
corrosion and build-up.

EOS has solved this problem by using a proprietary coating that creates a permanently conductive and non-corrosive surface.

Another start-up, Aquion Energy has raised more than $150M in equity and debt to deploy more than 1MW of sodium-
ion batteries at $300/kWh price point.

Aquion’s six-stack module, which is roughly the size of a refrigerator, can produce 10kWh, while its larger 100kWh cube module was recently deployed in Hawaii, where 40 units will be shipped during first quarter 2015.

Screen Shot 2015-03-03 at 11.43.12 AM

Reprinted with permission.

Tags: , , , ,

About the Author

is many, many people. We publish a number of guest posts from experts in a large variety of fields. This is our contributor account for those special people. :D

  • Bob_Wallace

    Sounds interesting Are you willing to disclose more details?

    • Christophe Stevens

      You can contact me directly through the contact email on the website. I can disclose more détails.

  • vensonata

    LCOE. The formula for establishing this must be completely taken apart by a few sharp people. The entire metric is suspect. Just do the lifetime production of solar panels without the arcane math called “the cost of money”. For instance in Phoenix, Arizona (best case scenario) Rooftop fixed at angle can produce 1750kwh per year from a 1 kw array. Installed price without subsidies can be had for $3.25. So multiply 1750 by 35 years (yes that is a mimimum true life expectancy with close to full production..but we shall assume 15% degradation in the last 5 years). Total kwh 61000kwh for $3250, = 5.3cents kwh. So there is the figure they somehow turn into 13cents by borrowing money at inflated interest rates and charging some imaginary maintenance mans wages to run a hose over the panels twice a year etc. Battery cost for a house means combining Pv and battery kwh only for night use of less than 40% of the daily total use. If people live at the modern standard of net Zero houses -about 20 kwh/day average total all electric. Then at overall cost of 12cents kwh plus PV/ Battery combined, they can expect to spend $876 year total. That is less than half the present average now on grid. So lets double everything and we are still less than average present costs. What is not to like. Do remember that 7000kwh per year total energy needs is realistic for a sensible 3 person family in a sensibly built house. Let us not bring in the senseless family in the senseless house.

    • Mike Shurtleff

      “Then at overall cost of 12cents kwh plus PV/ Battery combined”
      12cents kwh with PV/Battery combined?

      I’m sure they’re using average install costs for PV. Yes, it can certainly be done for less. Average install costs in US will drop from $4/W to less than $2/W in a few short years, as we know from Germany and Australia. They are already there.

      Storage Battery costs are going to drop more rapidly than they show, as I already stated. They already are.

      Both really make huge economic sense at the end-of-grid, makes going off-grid a huge threat to utilities going forward, especially in places with high electricity prices like Hawaii and Australia. This will drive change big time.

    • Aku Ankka

      It is true that choice of parameters strongly influences the cost.
      But I don’t think that omitting cost of money would make things fair at all — if monetary cost is the main metrics (need not be, but if so, it’s different discussion), you do need to consider that even if you do not have to borrow money (and pay interest), you could alternatively invest it and earn income (fixed or variable).
      So comparing solar production (with or without battery) you would compare it to regular buying from utility, and investing money elsewhere. Using cost of financing here is an attempt to level the playing field.

      Maintenance, too, is not some bogus add-on fee: if you need to do additional maintenance work, it should absolutely be included. Cost should be reasonably estimated, but it really is intellectually dishonest to claim it’s a non-existing thing, from economical perspective. You may well be happy to clean up panels, do minor adjustments if need be; and pay for occasional, rare (but existing) replacement of components that fail.
      These overall costs levelized over people with PV panels exist and can be estimated with historical data.

      As to size of household, usage, I don’t see how that has big significance for actual calculations; except that higher generation amounts are slightly lower cost as there is some fixed overhead. So math for higher consumption is likely to be slightly better for PV (price per W for installation bit lower).

      So I would rather see a rational, well-documented approach for doing real analysis, instead of trying to dumb down things to look more favorable.
      With the rate things are going you do not need to dumb things down to make the case for distributed solar. And doing that is begging for criticism from those opposed to the goal of more sustainable energy production.

      • vensonata

        That’s what I want to hear, critique of these various systems…including mine. You see I am simply not qualified to make definite assertions about how money works. Thats why I like to hear some critical responses.
        However I am heartened to find out that the Rocky Mountain Institute has considered three options for determining the actual value of PV production of electricity. (I quote from memory) “LCOE is instructive but not sufficient, LCAC (levelized cost of avoided electricity), is simple but not sufficient. And LTE (lifetime electricty production,) is perhaps the best parameter.
        So…beats me, but I do think that people who have far more analytical skills than me about Pv need to question the formulas which I suspect having been used because that is just convenient lingo. I might be wrong, anybody competent out there?

        • Aku Ankka

          I fully agree here. There are many ways to compare, and existing ones are probably not optimal. And at very least, more transparency is needed to lay out specific assumptions. While it is tempting to just give one number for simplicity, Einstein’s quote applies: “As simple as possible, but no simpler”.

  • We need to think in terms of a moon shot by the end of 1969. This is an excellent report so I’m not in contention. It’s just that “five years” sounds too elastic. I realize a bank analyst isn’t going to stick his/her neck out. But we should say, by January 20, 2017, the town of Cedar Rapids, Iowa will be completely powered by wind and solar, with 110 percent battery backup.” Nonetheless, I’m a big fan of mechanical battery systems. With all the scrap metal sitting around, why not build a flywheel the size of the original 1893 Columbian Exposition Ferris wheel? Form follows function. As they say in the arts and architecture world. Thinking small and light is essential for smartphones. But not necessarily the critical criteria for municipal/regional power systems.

    • Mike Shurtleff

      Friction with air is the problem with your giant wheel. Safety is another. Production cost is a third (needs to be factory production for lower cost). Sorry.

      Unimportance of weight for a stationary battery system is one of the reasons Ambri LMB looks to be a disruptive storage option to me. Deep cycle life is stratospheric, so LCOE for added cost of LMB storage will most likely be well below 1c/kWh. Combine with PV panel costs that are 40c/W in a few years and Kaboom! I expect the Solar PV + Storage market to explode before 2020.

      • So now what do I do with it? I just built it. So, I’ll build a giant housing to surround it and have it spin in a vacuum, like Volvo’s KERS system. Now take apart that brilliant modification, mister idea man. What is an Ambri LMB? Why do I want one. Do you work for Ambri LMB. Can I trust your opinion if you do? I’m too lazy to look it up. And I’m playing comments section engineer for fun.

        • Bob_Wallace

          Liquid metal battery. Surely you’ve read some of the articles about LMB?


          Why do you want one? If you’re a utility you will want a bunch (if they finish proving themselves) because they should be cheap and offer a 300 year lifespan with unlimited cycling.

          It looks from the table above that flywheels would provide the same lifespan. I wonder why more companies are not developing flywheel storage. Might it be that they don’t store as much power per volume?

          • I thought it meant Lazy Mike Berndtson. A big ol’ battery would be fine. Git r’ done. I still think all the folks interested in renewables should take one US city and turn it completely into a fossil fuel free power city. All electric generated by renewables. All cars and trucks. Heat. Etc. This kind of crazy thing was done to get cities to buy into new technologies in the past. For instance Sam Insul (Edison Elect. and Com Ed) built an enormous coal fired power plant before there were lights and toasters. This turned Chicago electric quick. And spurred much of Chicago’s manufacturing base. No, I’m not promoting coal (for the more literal folks out there). Just an example of how crazy ideas become not so crazy. And no again, I’m still not going to jump onto team Musk.

          • Mike Shurtleff

            AMBRI – Mass
            production to start in 2016. Prototype Production in Massachusetts has been up and running since November 2013. Prototypes have in under-going testing in installations. The chemistry as been improved. This one is amazing! Extremely low-cost over life of battery because of extremely high cycle-life. Competitive with pumped hydro power storage, without the geographical limitations.

            Cost has not been announced, but this battery has been design to use
            VERY cheap materials to begin with. I’m going to guess way high $400.00/kWh for purposes of calculation here.

            Deep cycle fade it 0.0002% per cycle. This would be only 1% degradation after 5,000 cycles, only 10% degradation after 50,000 cycles, only 20% degradation after 100,000 cycles.

            (50,000 cycles at 1 cycle per day = 137 years) (100,000 cycles at 1 cycle per day = 274 years)
            “Sadoway’s team says such batteries could last 300 years”
            Liquid electrodes are going to be self healing.
            (40000c per kWh / 50,000) = 0.8c/kWh cost of storage (..and I’m guessing high on cost)

            http://www.greentechmedia.com/articles/read/is-this-ambris-new-liquid-metal-battery-materials-formula – October 2014
            “Is This Ambri’s New Liquid-Metal Battery Materials Formula?”

            “The Electricity Grid Is About To Change… A Lot!” – October 2014
            Two very good short video clips of Donald Sadoway talking about energy in 2064.

            http://theenergycollective.com/jessejenkins/2142871/future-energy-will-cheap-dirt-batteries-transform-grid – October 2014
            “The Future of Energy: Will ‘Cheap as Dirt’ Batteries Transform the Grid?”
            Same two short video clips of Sadoway AND three good Grid, Solar, and Wind graphics (two shown below).

            http://reneweconomy.com.au/2015/australia-become-manufacturing-hub-battery-storage-70693 – January 2015
            “How Australia could become manufacturing hub of battery storage”

          • Bob_Wallace

            With both Ambri and Alevo talking about low manufacturing costs and very long cycle life we could have storage for less than a penny per kWh on a daily cycle basis.

            Add in 3 cent wind and solar (seems to be where we are heading) and the cost of “daily” electricity will be cheaper than what we pay now at the wholesale level.

            What I’d like to see is someone work up a model how far out battery storage could replace gas peakers.

            What I mean by that is as we store for more days (2 and up to two weeks) the price per cycle increases. But as well, the less a peaker plant runs the more it costs to produce electricity with it.

            A 0.8c/kWh single day of storage would likely cost about 1.6c/kWh with cycling every two days on average. 3.2c/kWh for four day average cycling. Where’s the crossover point at which gas would be cheaper?

        • Mike Shurtleff

          There are flywheel storage systems. I have some links if you’re interested. Sorry about the your ferris wheel idea, couldn’t resist responding to it. Didn’t figure you were serious. 😀

          see comments above on Alevo dark horse. If Ian is correct about $100/kWh then they could be even lower cost than Ambri LMB.

          see comment below on Ambri information. Very very low cost. I see no reason Ambri LMB could not be used for micro-grids, or for individual houses. It’s liquid metal, so hot, and I would not want installed in my house. Would have to go in cement building near house to prevent fire in case of earthquake. Ambri does not seem to have any current plans for household distributed storage. Just say’in it could be done. …and again very very low-cost of storage.

          Maybe Alevo will deliver at $100/kWh and do this. The better market is at the end-of-grid. Big price advantage there …and no utility interference.

  • Mike Shurtleff

    “Currently, according to Deutsche, the cost of a typical lead-acid battery may be as low as ~$200/kWh, while best in class lithium-ion technology was producing commercial/utility packages in the ~$500/kWh range at end 2014”
    …and not one of the plot lines in the “BATTERY PRICE PROJECTIONS” graph above is below $550/kWh for 2014, and only one is below $500/kWh in 2015.
    Elon Musk claims Tesla will reach costs of $150/kWh for their lithium batteries by 2018. Do anyone really think total battery cost will be over twice that, as shown in that graph.
    Costs may converge to $100/kWh over time, but they are going to come down much faster in the next few years. EIA prediction is just silly, like usual. Anybody take renewable energy predictions from the EIA seriously any more? Boy I’m glad I’m not one of those numb-skulls.

    • Ian

      In defence of the graph, it is a chart of the “blended” cost of the various technologies. I think that’s why the numbers are high- they might include some very expensive batteries. For me the surprising number is in fig 48- Emerging Companies. It says that Alevo’s battery is only $100/kWh! I’ve read elsewhere that it is capable of at least 40,000 discharge cycles with little degradation. I’d like to find out more about that.

      • Bob_Wallace

        Alevo has pretty much operated in stealth mode. I’m holding back on believing they are for real until they show us something. I’ve heard the low price claim, hadn’t seen the cycle life claim.

        Digging around I found an interesting article that starts –

        “Imagine 1000 gigafactories – that’s what we’ll be seeing in the coming decade.”


        Alevo claims their batteries can be charged in 30 mins and discharged over 40,000 times. And they seem to have rounded up a very large amount of money to get them going. Talking about being in manufacturing mode in 2017.


        • Ian

          Wow, I hadn’t heard that 30 min charge time! Alevo seems like it might be the real thing. Sure they’ve never gotten out of ‘stealth mode’- much like all the other failed batteries, but they have gotten a lot of investment money (half a billion), have bought a very large factory, and placed orders with outside suppliers. It seems like they know they have something ‘disruptive’, but no need to show their cards yet.

      • Mike Shurtleff

        Thanks for the info. I have information on Alevo that claims 40,000 or 30,000 cycles, but no info on cost. If thay can really sell

        In re-reading my comment I see I’m being harsh to focus just on the graph. Sorry about that. General text of the article is excellent. 2c/kWh in five years seems very reasonable to me. There are a few storage battery companies now in production and soon to hit the market, who could meet that price point. (eg EOS and Ambri, maybe Alevo too) It is likely they will actually sell at a premium over their manufacturing costs for a few years first.

        Excellent article! Yes, the graph is probably off some. Thanks for the response.

        • Mike Shurtleff

          $100.00 kWh / 40,000 cycles
          = 0.25c per kWh
          A quarter of a penny per kWh
          Boy, if that turns out to be correct it will be amazing!

        • Philip W

          I don’t think they will sell at a premium, since they all want to snatch as much market share as possible while the market is still relatively small. The market is going to explode over next few years. The more market share you have, the better.

          Oh boy, I’m so excited! 🙂

Back to Top ↑