James Montgomery, Associate Editor, RenewableEnergyWorld.com
February 05, 2014 | 3 Comments
Funding for Primus Power's Flow Batteries
Another flow battery company, Primus Power, has secured $20 million in a first close of Series C funding, led by one of its supply partners, South Africa-based Anglo American Platinum. That's a big step-up from the $15 million it had raised in two rounds from 2009-2011, plus some grants from the DoE, ARPA-E, Bonneville Power Administration, and California Energy Commission. CEO Tom Stepien noted Primus' batteries use a small amount of Anglo American Platinum's metal as a catalyst in the electrode, so adding them as a backer is strategic for both sides. Much like the market for catalytic converters in cars, "this could also be a beautiful market for them," he said.
One of the biggest elements in the flow battery cost stack is the separator/membrane, a polymer barrier to separate the liquid components of the system and shuttle protons through. Primus' architecture eliminates the physical ionic separator and instead utilizes a flow control system, not unlike those that manage some key semiconductor manufacturing processes, explained Stepien -- who would know, given his 13 years background at chip tool giant Applied Materials prior to joining Primus in 2009. Eliminating that barrier, in theory, leads to a big cost reduction. The system also uses a metal electrode, not a plastic and felt electrode, which is more expensive on an area basis but runs on 10× the current, and with 10× the power even at double the cost, "we win by a factor of five," he said. He also heralded the company's implementation of a single pump and a single tank (others have two of each), and a single set of piping, all of which reduce system costs and increase reliability, he said.
Also in the cost-reduction theme, Primus is contracting out its manufacturing, including assembly and testing. Stepien said a decision would be reached "in the next several months" among several candidates, all of whom have facilities in Silicon Valley. (Flextronics is one of them, he acknowledged; the company already has deals to make other companies' batteries, including fellow flow battery company Imergy.) The plan is to establish a "copy-exact" set of processes -- another takeaway from the semiconductor world -- and duplicate that elsewhere in the U.S. and "in a reasonable timeframe, other parts of the world," he said.
Primus is targeting large grid-scale stationary storage, in the form of investor-owned utilities and municipal utilities, and microgrids. The company will officially ship systems later this year to Puget Sound Energy (taking the place of a new substation) and to the Marine Corps Air Station in Miramar, California (part of a microgrid setup), both of which were announced last year. Primus' partner in the aforementioned DoE grant, the Modesto (California) Irrigation District, is also progressing; the plan is to put 25 MW/75 MWh of the company's EnergyPods throughout the site's 45-substation network, though actual system size is still being worked out. He also cited "several municipal utilities mostly in California" that are expressing interest, including a letter-of-intent from one of them. The commercial/industrial segment, e.g., distribution centers, is also attractive as an application to reduce demand charges, he said -- "shav[ing] the peak and fade the demand charges pencils out very nicely."
Primus was ahead of the pack (technically the second wave of flow battery developers) in reducing costs, getting around $600/kWh around five years ago, but li-ion has continued to keep pace with that benchmark to erase the cost-value proposition for flow battery tech, noted Sam Jaffe. Flow batteries still have advantages in long-term energy storage (4-10+ hours), but today's market the ROI is in shorter-duration applications such as frequency regulation and voltage smoothing: "they need the power, not the long-term energy storage," he said. Prices will continue to come down dramatically, creating gigawatt-level markets and regulatory changes around them -- as we're starting to see in California and to a lesser extent in Canada and Long Island -- but a broader addressable market is still several years out, he suggested.
Penciling out costs is part of the equation -- but can the tech work as efficiently and/or as long lifetime as promised? Stepien took the reliability issue head-on, noting that Primus has had "third-parties look over our shoulder to really make sure we're not being premature about the robustness we need." That includes taking the system to a local utility this fall (PG&E's San Ramon facility) before going to Puget Energy. "It has taken more time, but ultimately we think we have a more solid product," he said.
Note that this is the "first close" of a Series C round of funding -- Stepien acknowledged that the company will need to add to that within the next year, a "similar-size amount" to the ~$35 million raised so far. "We'll need about that much to be cash-flow breakeven," he said.
Next: Green Charge Networks, Why Storage and Solar Behind the Meter Works