Is Lithium-Ion unassailable in stationary energy storage markets?

Troy Ault

CTG’s Power Breakfast series is supported by Silicon Valley Bank and Wilson Sonsini Goodrich & Rosati. Thank you to our sponsors!


At our latest Power Breakfast this past Wednesday, we convened a great panel and room full of energy storage industry players, from start-ups and investors to utilities and project developers. Our panel consisted of:

  • Mark Perutz from DBL Partners, one of the first venture investors in Tesla
  • Ilan Gur, former founder of lithium-ion battery company Seeo (acquired by Bosch), Tech-to-Market director with the US Department of Energy’s ARPA-E program, and now founder and Managing Director of the Cyclotron Road incubator at Lawrence Berkeley National Lab
  • Katherine Ryzhaya, Chief Commercial Officer of Lightsource Renewable Energy, a solar-plus-storage developer, and former executive at Advanced Microgrid Solutions
  • Mateo Jaramillo, an entrepreneur-in-residence at Cyclotron Road and former VP of Product at Tesla
  • Pulakesh Mukherjee, a Principal with BASF’s venture capital arm
  • Tom Stepien, the CEO of flow battery company Primus Power

Our lofty goal was to address the question: is Lithium-Ion unassailable in stationary energy storage markets? We came out of the morning with some great perspectives from panelists and attendees alike. And, while we stick to our normal Chatham House Rule dictating non-attribution of such perspectives, we will attempt to distill back to you here several of the key learnings of the day.

Setting the stage

To start, we set the stage with the need for stationary storage and addressed the question of why lithium-ion is the dominant platform at this early stage. There are many well-documented possible benefits and markets for stationary energy storage, but our group boiled the need down to flexibility. Flexibility in energy pricing is the core benefit in use cases including customer side arbitrage, utilities can use storage to more flexibly accommodate intermittent renewable power generation and demand peak shaving, and grid operators achieve enhanced flexibility (better than even natural gas peaker plants can provide) in using storage for frequency regulation. Importantly, it’s worth noting that realists in the room helped to remind us all that lithium-ion’s main competitor – and a big competitor of any battery technology in stationary power – is natural gas power generation. The fast-ramping nature of natural gas power plants, the fuel’s current and sustaining low price, and the ability to store chemical energy in the form of natural gas in existing pipeline networks and on-site storage tanks make it a formidable competitor to electrochemical storage challengers.

As for why lithium-ion has achieved an early position of battery market dominance, the consensus seemed to be that growth in other markets like electric vehicles and consumer electronics has resulted in a more robust supply chain for lithium-ion than for any other platform. This supply chain advantage creates an element of inertia in lower cost manufacturing and availability of supply. In addition to supply chain advantages, though, a lot of the features that led lithium-ion to dominance in the electric vehicle and consumer electronics markets work well for stationary power, too. Cycle life and charge/discharge patterns match up with our diurnal patterns of energy demand, and dynamic discharge (lots of power all at once for high-torque, as well as steady, set-the-cruise-control, discharge over time) characteristics needed for vehicles tend to match up with various use cases on the grid.

Several in the room helped us understand that further strength is derived from the fact that lithium-ion is not one chemistry, but a flexible platform consisting of several possible chemistries and configurations. While Moore’s law from the semiconductor industry cannot be directly applied to batteries, what the market saw was a Moore’s-law-like doubling of performance in lithium-ion roughly fifteen years after its initial introduction by Sony, thanks to rapid learning cycles and slight tweaks to the materials that lithium ions are cramming themselves into in the battery electrodes. That can be compared to the same doubling in performance over more like 60 or 70 years for other battery chemistries like lead-acid and nickel-cadmium. Innovators today are still working at electrode improvements in lithium-ion, as well as solid-state electrolytes. All of this is moving the target even further away for competitors.

Competitors (besides natural gas)

As for competitor technologies, most investors and developers in the room opted not to name what they’d bet on besides lithium-ion, but we did hear from investors indicating current due diligence on other types of batteries.

Customers in the room emphasized the importance of a proven product on both performance and cost, and that they are otherwise chemistry-agnostic. In addition, however, stability of the supplier – will they be around years down the road when maintenance is needed on a 25-year utility contract – is a key element in buying decisions.

There are key openings where lithium-ion has weaknesses, however. Once you get to the 6-hour or longer duration market for stationary energy, lithium-ion starts to break down as the optimal solution. Long-duration storage solutions like flow batteries could have an opening here. Also, lithium-ion has an inherent issue with cycle-life degradation that leaves an opening for competitors. As charge/discharge cycles accumulate, the battery’s usable capacity degrades. This will be familiar to smart phone owners who find themselves having to recharge their batteries more frequently as the phone’s battery ages. Thus far, lithium-ion suppliers to stationary energy storage projects have mitigated this problem via long term maintenance agreements that guarantee no degradation at the plant level. So, they replace battery cells at a predictable cost and frequency that the aforementioned supply chain advantage positions them particularly well to support.

Competitors ultimately need to get the run-time hours necessary to prove new chemistries and technologies as bankable. The consensus in the room was that this is the key area where start-ups need to be scrappy – in finding edge cases and markets or partners where they can prove themselves and validate their technology for bankers, insurers, and bigger customers.

There was, however, also a recognition in the room of information asymmetries and bending-of-the-truth by battery suppliers regarding system costs, particularly at the early stages of development for new technologies. Ultimately, there needs to be a recognition that a single battery might have different end cost characteristics in different markets, and so suppliers may be hesitant to issue broad-based cost data. The industry could, however, benefit from a third-party validator of industry standard balance-of-system costs. It was pointed out that Sematech provided the same service in semiconductor manufacturing when similar challenges existed in that industry.

Aquion Energy

Finally, underlying the day’s conversation was the recent failure at the start of the year of Aquion Energy, a developer of one of the most promising lithium-ion competitor technologies thus far, using salt water. The room engaged in a robust discussion of why the company had failed, and ultimately came to four conclusions:

  • First, the company underestimated how quickly the cost of lithium-ion would come down.
  • Second, when you’re trying to introduce a new technology, it helps to be able to subsidize the launch via a large balance sheet and other revenue sources, and operate at a loss initially. Inherently, for a venture-backed start-up like Aquion, the only resource is raising more capital, which becomes hard as servicing debts and producing returns to equity investors create a strain on the business.
  • Third, there was a suggestion that partnership models for both manufacturing and deployment should have been the focus versus Aquion building, owning, and operating its own manufacturing capacity and projects. For any small company, management has to think about when it will build up the company’s manufacturing because you don’t want to lose business by not having enough manufacturing capacity, but you also don’t want to build too early and have your investment outstrip your revenue growth. Ultimately, Aquion failed to thread this needle.
  • Finally, several market experts seemed to agree that the company built a great, safe, 20-hour battery (a battery capable of supporting 20-hour duration of constant discharge). The battery was relatively easy to control, to boot. But the nail in the coffin may have been that there was simply not yet a robust market for a 20-hour battery. Such long duration markets are expected to develop, particularly as renewable energy penetration on local grids increases, but the technology may simply have been ahead of its time.


Look for future editions of our Power Breakfast series, advertised in our Inside Cleantech newsletter and on our Upcoming Events page. 

Tweet about this on TwitterShare on Google+Share on FacebookEmail this to someone
  • Once you get to the 6-hour or longer duration market for stationary energy, lithium-ion starts to break down as the optimal solution.

    For large capacity, grid-scale, energy storage, pumped-storage hydro is optimal at the 6-hour duration and can be competitive at 6 minutes where hydro power is in plentiful supply. Only at 6 seconds are batteries clearly optimal at any scale.
    5 or 6 hours of regeneration power is only sufficient capacity for energy storage to operate profitably in arbitrage mode, buying low price electricity and selling at a premium price.
    However, grid management really needs energy storage of 5 or 6 hours times solar or wind farm capacity and that’s a lot more energy storage capacity.

    Do the calculations to estimate the energy storage required to transform the intermittent output from wind power into a reliable generator with the Wind, storage and back-up system designer
    Peak demand, wind and back-up power / energy usage and storage capacity calculator

    For the specification and design of renewable energy electricity generation systems which successfully smooth intermittent wind generation to serve customer demand, 24 hours a day, 7 days a week and 52 weeks a year.

    Adopting the recommendation derived from scientific computer modelling that the energy storage capacity be about 5 hours times the wind power capacity, the tables offer rows of previously successful modelled system configurations – row A, a configuration with no back-up power and rows B to G offering alternative ratios of wind power to back-up power. Columns consist of adjustable power and energy values in proportion to fixed multiplier factors.

    Scottish Scientist
    Independent Scientific Adviser for Scotland

    * Wind, storage and back-up system designer
    * Double Tidal Lagoon Baseload Scheme
    * Off-Shore Electricity from Wind, Solar and Hydrogen Power
    * World’s biggest-ever pumped-storage hydro-scheme, for Scotland?
    * Modelling of wind and pumped-storage power
    * Scotland Electricity Generation – my plan for 2020
    * South America – GREAT for Renewable Energy

  • Mike Grunow

    So much group think going on.

    Will someone show me a Li-Ion project with a life over 5 years? Will someone find 1 human who has a positive thing to say about their cell phone battery? Will someone find me 1 utility executive who wants a potentially explosive battery next to a critical substation without a 300 foot setback?

    The real key here are the services that battery use cases pay for. They pay for power services and any challenger needs to be able to provide power and energy services. Aquion was an energy service battery only and the duty cycle on energy only service use cases is 50-60 days out of the year and owners want to use the batteries for high power ancillary services the other 300 days out of the year. I would not expect this group of Li-Ion disciples talk honestly about the challenges of Li_ion over 10 year contracts.

    Ask the controls guys and they will tell you that replacement cells 7 years out will not be 1 for 1 reverse compatible with current chemistry electrically and will require custom controls and whole-hog swap out.

    Next, please go talk to the owners of the ~300MW of merchant PJM projects that are facing stranded asset status with a simple change is signal. Li-Ion batteries are 1-trick ponies that are not versatile enough to provide power and energy from 1 platform.

  • Francisco M Pericão Faria

    “Aquion…technology may simply have been ahead of its time.”
    Really? good thing then it’s coming back to market…