As electric vehicles proliferate, so too will used EV batteries. Car companies and researchers are hustling to figure out how to safely adapt and reuse those depleted batteries when that time comes.
The basic pitch is simple enough: cars demand very high performance from their batteries, so once the battery’s capacity declines past a certain point -- 70 percent or 80 percent, depending on who you talk to -- it needs to be swapped out. At that point, though, the battery can still handle a lot of charge and discharge, making it useful for storage in less intensive stationary settings.
The sheer expense of developing and building those batteries in the first place makes a compelling case for capturing some additional value after their initial use. Second-life applications also delay the need to dispose of these resource-intensive products, which nobody has yet figured out how to do economically. If storage vendors can resell used batteries as a cheaper alternative to new storage, they could help more people consume their own rooftop solar generation, or reduce their peak demand, or any number of other uses that would advance the progress of a low-carbon grid.
This is new territory, so there are a lot of questions yet to be resolved. What are the engineering challenges involved in taking batteries from mobile use to stationary use? Will they perform as well in the new capacity? How do you standardize across different degrees of wear and tear?
There is also the matter of setting up markets around used batteries and determining who benefits from that trade. Before that can happen, the physics of the transition have to be worked out. GTM asked some people pioneering the field about what they’ve encountered so far.
How do you standardize quality from disparate used batteries?
Commodities markets rely on some degree of standardization to ensure that customers know what they’re getting for a certain price. New EV batteries have to meet exacting specifications, but the ones that come out of the chassis after years of driving will perform at all kinds of different levels.
“Batteries are a lot like people: They each have their own individual state of health depending upon what they've been exposed to and how they've been treated over the course of their life,” said Ken Boyce, who’s developing a safety standard for second life batteries at Underwriters Laboratories, a major safety certification firm.
Also like people, batteries are made up of multiple interlocking systems -- the health of the pack as a whole depends on the health of each of the modules, which depend on the health of the cells within them. Second-life developers have to account for differing wear and tear at each of these levels, in order to standardize the packs and put together storage units that perform at a consistent and predictable level.
Moreover, they’ll have to ascertain that information without wrecking the cells in the process. Noninvasive procedures are necessary for the batteries to remain useful after the assessment.
The good news is that methods for testing the batteries are not daunting. Evaluators can track rates of electrical charge and discharge, use thermal imaging to screen for abnormal performance, and parse data from the battery management system, which governs its functions.
Once the standard, known as UL 1974, comes out -- Boyce said it should be ready by the end of the year -- that will guide second-life battery developers in the level of quality they need to meet.
“Making sure that we've thought through that collectively as a technical community and captured those requirements in a standard is really important, so that as they go into that brand new application, everybody can really feel comfortable about that and expect that they will perform in a safe and sustainable manner,” he said.
How do you match up old batteries?
Meeting the UL standard won’t be easy. It requires studying the differences of each used battery and accounting for them in the stationary system design. But it’s possible, because it’s already been done.
The University of California at San Diego set up the first large-scale second-life battery installation back in 2013. It networked 100 kilowatts and 160 kilowatt-hours of batteries from BMW’s first EV, the Mini E, and connected it to the microgrid that powers almost all of the campus’ electricity needs.
“It took a lot of effort; no one had done this before at the time,” said William Torre, program director of energy storage research at UCSD. “We had a lot of challenges with the development of the control systems and getting them communicating properly.”
Once they’d figured out the battery management systems for each pack, they had to create a “super BMS” to oversee all the different battery management systems. The next key step was categorizing each of the batteries so they could group together batteries with similar states of charge and capacity.
“You can do some things with the control system to accommodate some variance in the battery packs and their voltages, but if you can keep them closely within range, you're going to get more use out of the batteries and more capacity out of the system,” Torre said.
The process for networking multiple batteries with different use histories is no longer a mystery. The resources needed to accomplish this, though, eat into the profits that companies will eventually earn from selling second-life battery systems.
That cost of doing business will likely come down as the process becomes more routine. EV manufacturers could help the process by standardizing their battery components, so second-life adapters have less variety to deal with.
“The process, the labor involved in that process, and the cost could impact the business case for mass use of second-life EV batteries, but I'm confident that we’ll be able to overcome that,” Torre said.
The design of the original EV battery can also minimize the expense of converting it for stationary storage, said Cliff Fietzek, manager of connected e-mobility at BMW of North America. Early BMW EV batteries were made up of several components that were then disassembled and rewired to make a second-life battery system. This cost more than BMW’s researchers had expected, so they redesigned things for the newer i3 battery.
The new batteries are self-contained units; when they come out of the car, they pull out with the housing, heating and cooling systems, and BMS intact. That makes it easier to service and replace the batteries, but also paves the way for use in stationary storage.
“We have seen that the less modification you have to do to the battery to integrate it into a storage application, the more cost-effective it is,” Fietzek said.
How do you maintain safety?
Pulling the batteries out of the automotive context for which they have been approved and setting them in a new environment requires a new round of safety precautions, even if the technology remains fundamentally the same.
“Whether it's in the belly of an electric vehicle or in a stationary energy storage system, the issues are the same issues, and it's really keeping those cells in that safe operating mode and making sure we're governing the processes appropriately,” Boyce said.
If anything, though, stationary storage should feel like a relaxing semi-retirement for the batteries. Car use exposes them to constant vibrations and repeated, sudden demands for power. In stationary life, the batteries just sit there, drawing in power from rooftop solar perhaps, and discharging when electricity is most costly.
“Probably it’s safer in your garage than it is in a vehicle,” Torre said. “In a car, of course, you could get into a car accident. It's moving. Stationary storage is just mounted on a wall in your garage.”
The batteries still need to come with redundant safety systems, but that’s par for the course. And, as BMW’s Fietzek pointed out, EV drivers are already accustomed to the concept of having an EV battery inside a building. “We are confident enough that you can park you i3 in your garage without your house burning down,” he said.
The home or business setting creates some new risks, just as any battery installation introduces. Fire departments are at work on new policies for responding to blazes that feature energy storage installations. There just doesn’t appear to be any elevated risk resulting from the prior automotive service of the batteries.
What will the market look like?
Once the fledgling second life-batteries make it through the physical and regulatory hurdles, they still have to find a buyer. The market landscape remains mysterious, because it will still be several years before the first mass-market generation of EV batteries starts to retire.
The general picture taking shape is a system in which the original manufacturers buy back their batteries from customers when they need replacement for automotive use. Then the company will repackage and sell them as a stationary storage system. BMW and Nissan, for instance, have already announced home storage products that use their old EV batteries.
It’s possible that a do-it-yourself culture will emerge where intrepid EV owners pull out their own batteries and tinker with them to produce something fun and useful. The dangers of playing around with expensive electrical systems will likely discourage most people from attempting this, and the manufacturers will no doubt try to limit the after-market manipulation of their products.
It might also be difficult to achieve much without access to the proprietary software that governs the battery’s internal communications, Fietzek said. He added that it’s possible large battery integrators will be able to source from different EV manufacturers, provided they group batteries with similar remaining capacity.
BMW is planning some demonstration installations in early 2017 to learn more about the market for residential storage, like what the customer is willing to pay, what the use cases are, and whether utilities are interested in partnering to aggregate second-life storage for grid services.
In markets with high demand charges, a strong use case already exists for pairing second-life batteries with EV charging stations to lower monthly demand charges. EVgo, which provides DC fast charging for electric cars, has been testing this model in a pilot at UCSD since earlier this year.
“Demand charges at DC fast-charging sites are a big liability in terms of operational costs,” said Niki de Leon, distributed generation program manager at EVgo. “As soon as somebody plugs in -- boom, you’re charging at 50 kilowatts. If multiple customers show up at the same time and all those demand charges are stacking up, even if it only happens once a month, then we’re stuck with a very high utility bill.”
The pilot results have been favorable, and the company now plans to deploy at two off-campus sites before the end of Q2 2017, de Leon added. The key business drivers for using second-life EV batteries are the lower cost and the opportunity to partner with the car manufacturers.
“The big takeaway is, it’s working and we’re really excited about it,” she said. “We’re seeing very favorable payback periods.”
How will second-life batteries compete with new ones?
As for who will buy the batteries, that remains to be seen. As Vox’s David Roberts pointed out recently, used batteries won’t be operating in a vacuum: They’ll be competing for market share with all the new batteries that will, in all likelihood, be much cheaper and better than anything on the market today. The fact that a lot of used EV batteries exist doesn’t guarantee that people will want to buy them.
That said, we still lack a method for recycling them economically; disposal costs a lot more than the value of the disassembled parts. As long as that remains the case, manufacturers will have a financial incentive to repackage their batteries and eke a little more value out of all the R&D investment that went into them.
Perhaps second-life batteries could play a role in spreading storage to new customer classes. New clean-energy products have historically cost too much to be accessible to low-income customers, who are more likely to live in neighborhoods with higher levels of air pollution and other environmental concerns. Home battery systems currently fall into that costs-too-much category.
Used batteries, then, could create a shadow economy, much like the used car market that exists today: if you can buy new, you’ll get the latest technology and the shiniest package, but you can save a lot by buying a used product that still retains all the core functionality you actually need.
Another factor in favor of a used EV market is that stationary uses have less constraints around space and weight to worry about. Lots of R&D dollars are going into new batteries so they can take cars farther without using up more space -- increasing the energy density, in other words. A stationary storage system could simply pack in more used batteries and achieve the same effect. That’s easier to do if, like BMW’s batteries, they come out of the car with all the needed cooling and safety systems intact.
And if the market assigns a much lower value to used EV batteries compared to new stationary storage, that wouldn’t be such a bad thing for second-life storage, Torre argued. After all, second-life storage developers aren’t selling a commodity so much as a service.
“That would be a great business to get into, where you can buy the equipment for free and install it and make revenue,” he said. “I doubt that it will ever get down that far, but I think as we scale up...mass-producing more batteries, the hope is that energy storage costs will come down.”