Back in February, Tesla Motors sent a shock wave through the energy technology world when it announced plans to build the globe’s biggest battery factory.
The sheer scale of the proposed “gigafactory” is enormous. “[Tesla’s] goal by 2020 is to be producing 500,000 cars — 500,000 battery packs — out of that gigafactory,” said Steve LeVine, a journalist for Quartz who’s writing a book on batteries and their potential to transform energy as we know it. If Tesla hits that target, it would literally double global production of lithium-ion batteries.
For anyone concerned about climate change, this is big news. Making every vehicle electric would reduce carbon emissions by moving cars off a pure oil diet and onto an electricity mix of coal, natural gas, and some renewables. But we need to move that electricity mix fully onto clean energy as well — and do it fast — to avoid the worst impacts of climate change.
That will require dealing with the intermittency of green energy sources like wind and solar. Electricity is relatively unusual in that we don’t make it until we use it — and if we don’t use it right when it’s made, we lose it. So fossil fuels like coal and natural gas have an advantage in that we control when we burn them. But the sun shines and the wind blows where and when they will, limiting renewables to a supplemental energy source at best. To change that, we need to be able to store renewable electricity when it’s made, and then release it when we need it.
And that means batteries. Lots and lots of batteries.
Changing The Game
But Tesla’s gigafactory also reveals the big hurdle still standing in the way of batteries: cost. As the size of a factory increases, the number of units of product it can produce — batteries in this case — goes up faster than the total costs of the equipment and staff and everything else that makes the factory run. A bigger factory and more production will bring the cost per unit down, allowing Tesla to cut the prices it charges the consumer.
In other words, Tesla hasn’t hit on a technological solution to batteries’ costs. They’re just trying to mitigate those costs through brute economic force.
This wasn’t what many were expecting in 2009. The stimulus bill pumped hundreds of millions of dollars into cutting-edge energy research at United States tech firms, and into labs like Argonne National Laboratory and the Department of Energy’s ARPA-E agency. The aim was to finally secure the breakthrough that would make battery technology ubiquitous in the energy economy, “to develop the next generation of fuel-efficient cars and trucks powered by the next generation of battery technologies all made right here in the U.S. of A.,” as President Obama put it. LeVine himself started his book in 2011, “pretty wide-eyed,” only to see the expected breakthroughs stall and several major battery firms go bankrupt.
“I was fooled, as I think a lot of us were, by the era we grew up in,” LeVine mused. “Hearing about Apollo and Kennedy with the ten year goal, and we know about the Manhattan Project, and we also have Silicon Valley on our brain and the internet and all that. So we think all you need to do is have a leader declare great goals, and then you throw money at it, and in ten years you’ll have it.”
“It turns out those examples are aberrations. And the really hard things are really hard.”
Ever since Alessandro Volta built the very first battery around 1800, the basic concept behind making batteries hasn’t changed. We’re still using a chemical reaction to drive ions between two oppositely charged materials, which in turn drives the flow of electricity. But the materials themselves keep changing as we try out new battery “recipes.” Volta had zinc-copper; the last hundred years saw lead-acid batteries; more recently there was nickel-metal hydride, which is in the Toyota Prius and most airplanes; and finally we hit on lithium-ion, which is what powers everything from our laptops, our smartphones, and our cordless power tools, to Tesla’s Model S and General Motors’ Chevy Volt.
When it comes to batteries these days, the goal is a better rate of improvement in performance: more stability and more energy storage for less cost. Each recipe has delivered a better rate, and the lithium-ion batteries Tesla’s gigafactory would produce are the best yet. “Since 1991, you’ve been able to increase the amount of energy you can put in that [lithium-ion battery] roughly five percent a year,” said Jeff Chamberlain, the Deputy Director of Development and Demonstration for the Joint Center for Energy Storage Research at Argonne National Labs. “And at the same time reduce the costs by about seven percent per year. And that’s really good.”
The car companies are betting that battery scientists will fail for the next three decades.
But it’s not enough. Chamberlain compared it to microchips, which have been doubling their performance every 18 months. And not through any change in materials — they’re still just patterns of silicon conducting electricity — but through perpetual improvements in design sophistication and manufacturing methodology. Not even lithium-ion batteries have gotten anywhere close, and they’ve been around for two decades.
“If you look at the car companies, natural gas companies, oil companies, their publicly released forecasts for 30 or 40 years down the road show that electric cars will still only have one percent of the car market,” LeVine said. “They’re betting — it’s really a sounding — that battery scientists will fail for the next three decades. Which is incredible!”
Innovating More Efficient Vehicles
For the moment, that leaves everyone in the automobile industry trying to get the best they can out of existing technology.
Tesla’s own efforts don’t end with the gigafactory. Smaller batteries are more stable, so instead of one big battery Tesla actually uses a pack of a few thousand small ones, all just a little bigger than your average AA battery. They’re linked together by Tesla’s “management system,” which coordinates the batteries to get the charge the car’s electric motor needs, when it needs it, and in the safest and most efficient manner.
No one really knows how much the battery packs cost, and Tesla obviously isn’t saying. But Chamberlain estimated it’s a substantial portion of the car’s $69,900 price tag. (Before sales taxes, tax credits, etc.) According to LeVine, Tesla CEO Elon Musk thinks the gigafactory’s scaled up production will cut the cost of the battery packs by 30 percent, which could reduce the overall cost of the Model S considerably.
Meanwhile, General Motors has decided the best way around batteries’ current limitations is to keep an internal combustion engine as a back-up. The Chevrolet Volt’s lithium-ion battery has only a 40-mile range (versus the 180 to 260-mile range on the Model S) but if the driver runs the battery down, the gasoline engine kicks in to keep generating power for the electric motors. GM went with such a small range because, after pouring through a wealth of Department of Energy studies on electric vehicle use, they found that 78 percent of Americans put in a round-trip commute to and from work of 40 miles or less per day. “If we can get almost 80 percent of Americans off of gas every day during their commute, that’s the target,” said Britta Gross, a former member of GM’s research and development team, who now heads their infrastructure development. “Just remove emissions from the tail pipes of vehicles but not constrain consumers.”
The research also showed drivers were actually clocking just as many, if not more, all-electric miles in the Volt as they were in other purely electric vehicles. The all-electric drivers, according to Gross, are nervous about running their battery down past even half capacity, given the uncertainty of finding somewhere to recharge. “Volt drivers are rigorous about not trying to dip into the gas,” Gross explained, but thanks to the gasoline-powered back-up “they’re comfortable using all of the battery capacity.”
That logic allows GM to use a smaller battery, which means a much cheaper battery, which gets the Volt to just over $34,000. Throw in the federal tax credit, and it’s within spitting distance of the Toyota Camry — the most popular four-door passenger car in 2013.
The Military Option
When it comes to the electricity grid, batteries face the same challenge as on the car market: reduce costs and improve energy storage and performance. But there’s no real equivalent workaround on the grid, and the applications multiply wildly.
To get an idea of what the future of this might look like, we can actually turn to the military. Back in 2005 or 2006, the military suddenly realized it had a big problem with how it was powering its bases in Iraq and Afghanistan. “You would have one 60-kilowatt generator hooked up to something that needed 10 kilowatts,” explained Michael Wu, the Energy Program Director for the Truman National Security Project & Center for National Policy. “So you were wasting an incredible amount of fuel because you weren’t linking together generators.”
Beyond that, the generators required extensive battlefield supply lines to get fuel to the bases, and that put soldiers at risk. “We lost a lot of lives in Iraq and Afghanistan trying to protect those convoys,” Wu said. So the military started replacing those generators by investing in solar in a big way — solar tents and solar blankets and tactical solar panels and other forms of easily-deployable solar power. Combine that with batteries, and warzone bases could eventually generate all the power they need on-site, without the need for fuel convoys.
Beyond that, soldiers and marines on missions could carry up to 18 different types of batteries to power their laptops, communications equipment, and other electronics. “Our biggest tactical advantage over the enemy is that we have unparalleled command and control,” Wu said, “which is only possible through our communications equipment.” But those batteries weighed the soldiers down, so the Defense Department streamlined it all into one integrated battery system. The military also designed a wearable vest — the Marine Austere Patrolling System — that features both solar panels and an attached battery that connects to equipment via USB technology. It even has a water filtration system.
State-side military bases “are directly supporting the war fighter in operations around the world like they’ve never done before,” Wu added, pointing to the aerial drones that provide soldiers in Afghanistan with reconnaissance and intelligence, and which are almost all piloted remotely from the states. “But right now the military is dependent on the civilian grid for 99 percent of its electricity requirements,” said Wu. “And that civilian grid is fragile and nearly a century old. In fact in 2012 alone, our military installations had 87 power outages of eight hours or more. And that’s unacceptable.”
That instability is driving a domestic move towards renewable energy for the military, as well. Some of it is coming from wind power on military-owned lands, but most, again, is solar — ultimately aimed at making bases’ energy production entirely self-sufficient. At the Marine Corps Air Station in Miramar, California, for instance, Primus Power is providing modular batteries that can be stacked together, reminiscent of Tesla’s battery pack, to provide storage for the base’s solar array.
Vehicles on military bases are also going electric. They tend to stick to low speeds and to run regular routes, ideal for cost-effective battery-powered motors, and they return to the same spot each night, which makes charging easy. The military has also started using the batteries in those vehicles as ad-hoc storage when they’re plugged in, smoothing out changes in electricity demand on the bases, and for the larger civilian electric grid. They’re turning to microgrids — computer systems that coordinate supply, demand, and storage of electricity, while keeping track of available battery capacity — to manage those electric vehicle fleets and bases’ electricity needs more broadly.
This all relates back to the civilian world in two ways. First, the military is a reliable customer with a lot of disposable income that can give private sector start-ups the support to bring their technologies and services to commercial scale. “Companies like Primus Power are benefiting from test-bedding to prove their technology and bring down their price, so that they become really attractive to the civilian sector,” Wu explained.
Second, all of the ways the military is using batteries can be mirrored in the civilian world. Pilot projects are experimenting with microgrids for homes and businesses to manage their own electricity use and storage, and to bring in electric vehicles as additional battery storage capacity when they’re plugged in at the garage or the parking lot. Delivery and shipping companies in cities are turning to electric vehicle fleets for the same reasons as the military: regular and predictable routes, low speeds in urban traffic, and regular returns to the same charging site. Companies like SolarCity are already providing home owners and office buildings with deals to hook up their solar arrays to Tesla-built batteries. “When I’m at work and I can collect sun energy at my home,” Chamberlain asked, “if I’m not using it at my home — where does it go? Wouldn’t it be great if I could store that energy in a small cheap battery in my garage, so when I get home and the sun’s down, I have sunlight right there in the can.”
Gross even mentioned that GM has been talking to utilities about repurposing used batteries for modular storage. “Some of these utilities are really faced with footprint problems where they can’t just put big rooms full of batteries at the base of a wind turbine or in the middle of Manhattan,” she said. “So that would be a case where a lithium-ion battery coming off a vehicle could be a good cost equation for them.”
The ultimate goal is a decentralized network of coordinated, overlapping battery storage; both large and small; at homes, businesses, skyscrapers, and utilities; throughout the country’s electrical grid. “What we’re aiming at,” said Chamberlain, “is transforming the grid so there’s this buffer zone of [stored] electricity.”
Helping The Revolution Along
What we’re aiming at is transforming the grid so there’s this buffer zone of [stored] electricity.
Americans’ demand for electricity goes up and down all the time, sometimes predictably, and sometimes unexpectedly and dramatically. That requires both “baseload” power — the minimum amount of electricity that always needs to be available on the grid — plus rapid and high frequency responses from utilities. Right now, that’s basically all done with natural gas turbines and coal burners run on complex computer algorithms to anticipate demand. Providing baseload power and meeting the demand changes are services batteries need to step up to provide.
There’s also the inevitable threat of blackouts. On the dramatic end, Hurricane Sandy left hospitals and multi-family residences in New York City without power for days — usually in the city’s poorest areas. There are everyday blackouts in the United States, as well, though they’re rare enough that buying batteries as backup isn’t a cost-effective choice for most households. But in other up-and-coming countries like India and China, where the grids are older and blackouts are more common, it’s not unusual to see a middle class family with a stack of batteries in their garage. And of course, there’s the need to smooth out the intermittency and unpredictability of wind and solar with better storage.
But the ugly truth is that the carbon-intensive, energy-inefficient way we handle all those problems now is still usually cheaper, in raw economic terms, than investing in batteries and the clean and coherent approach to energy they can provide. There are things we can change in the policy environment to help: more government-funded research, for one. Reforming regulation of the grid would help us get closer to an actual market in electricity, with individuals, along with utilities, making and even selling their own power. Passing a national cap-and-trade system or a carbon tax would build the future damage of climate change into the price of fossil fuels. That would make renewables — and, in turn, batteries — more economically attractive.
As for electric cars, Gross pointed out that the overwhelming majority of time people spend charging their cars is at their homes at night or at their work parking spot during the day. The Volt can already be fully recharged by plugging into a standard 120-volt, 3-pronged household socket for 10 hours. So we need new laws, building codes, and infrastructure to make sure every American has a plug with a dedicated circuit available in their home garage, in their office parking space, or in their apartment parking lot.
The Future Of Batteries
What we need to be transformational for the automotive industry and the grid is about another four or five-fold improvement.
In the end, technology is the key. “Our batteries today hold about six times more energy for a given structure of mass, compared to over 100 years ago,” Chamberlain said. “But what we need to be transformational for the automotive industry and the grid is about another four or five-fold improvement.”
First, we need better performance out of the battery “recipes” we already have. Unlike a microchip, which boils down to electrons moving through metal, a battery works through an actual chemical reaction. When it recharges, that reaction is reversed. To be worth it to the consumer, that battery has to be able to cycle through those reactions thousands of times over. That comes with consequences: over time, the physical make-up of the materials in the battery changes. Things warp and expand and shift, unexpected side reactions occur, performance degrades, and eventually the battery dies.
“To really invent the perfect battery, you want to understand every single aspect of that chemical reaction,” Chamberlain explained. “So now, in the last decade or so, we actually have the tools to do it. We have supercomputers and we have X-ray sources and electron microscopy, where we can actually watch materials all the way down to atoms and see what is happening.”
That means building new materials, from the very bottom up, through nanotechnology. One example is a battery’s electrolyte, which sits between the electrodes — the two oppositely charged ends of the battery — and facilitates the movement of the ions between them. Electrolytes are typically a liquid chemical, but if they can be changed to a nano-engineered solid, the life and stability of lithium-ion batteries would rise significantly.
Another example is new materials for the electrodes themselves. We use carbon now, but if we could move to silicon, lithium-ion batteries could store much more energy. Unfortunately, silicon also reacts to lithium by expanding around 300 percent, so we need to nano-engineer a containing structure that can take thousands of repeated expansions and contractions. “We’re using graphene, which is a relatively new structure that was discovered in the last decade,” Chamberlain said. “What’s great about that is it’s incredibly [electrically] conductive and it’s very flexible.”
To really invent the perfect battery, you want to understand every single aspect of that chemical reaction.
“Lithium-ion at its theoretical best will be somewhere between two to three times better than we are today. All of that stuff is being worked on. Not just at Argonne or Berkeley but literately all around the world. And it’s quite likely we will get there,” Chamberlain said.
Then there are technologies that redefine our idea of what a battery is. Flow batteries, for example, have liquids instead of solids for their electrodes. They’re recharged by switching out the tanks, and how much energy they store is determined purely by the tanks’ size. (That’s what Primus Power is using at Miramar.) Or pumped hydro storage, in which the rechargeable battery is literally two huge reservoirs of water at different heights, with pipes and turbine generators in between. The projects are only cost-effective as long as the landscape itself provides the downhill drop for the water, so they’re limited by geography. But they hold the potential for massive large-scale energy storage.
But getting all the way to five times better will require combining those new nano-engineering techniques with all new recipes that expand or even move beyond the lithium-ion umbrella. Things like lithium-oxygen, lithium-sulfur, or sodium-sulfur batteries for homes and cars and even utilities.
“The problem is the physics and chemistry behind those reactions is significantly more difficult to understand and control,” Chamberlain explained. “But they will very clearly get us to four or five times from where we are today.”