Berkeley Lab battery scientist Gao Liu inspects coin cells cycling in an environmental chamber. (Photo by Roy Kaltschmidt, Berkeley Lab Public Affairs) |
With
several new models of electric vehicles hitting the market this year
and more next year, President Obama’s goal of putting 1 million EVs on
U.S. roads by 2015 is tantalizingly within grasp. But what will it take
for that number to reach 10 million or even 100 million in 20 years?
The
answer: batteries need significant improvements. Specifically, they
need to be cheaper, safer, last longer and have higher energy. The
battery research team at Lawrence Berkeley National Laboratory (Berkeley
Lab), recognized as one of the best in the country, is engaged in
high-risk, high-reward research in each of those four areas, striving
for technology breakthroughs as well as incremental advances. Their work
could help drive a transformation of the vehicle industry and make EVs
as common as laptops and cell phones for American consumers.
“I
think with incremental improvements in batteries, engineering advances
in the car and support from the government, these are all things that
will make it a reality,” says Berkeley Lab scientist Marca Doeff. “And
there’s considerable enthusiasm among the population as a whole, so I
think it’s going to happen.”
Indeed,
it is a boom time for batteries. In the last three years, the battery
group at Berkeley Lab has hired 24 researchers, and the budget for the
Department of Energy’s Batteries for Advanced Transportation
Technologies (BATT) program, which is managed by Berkeley Lab, has grown
from $5 million four years ago to $16 million this year. More recently,
Berkeley Lab’s battery team was part of two multimillion-dollar awards
from DOE’s Advanced Research Projects Agency-Energy (ARPA-E) funded by
Recovery Act money. In one, the Lab is working with Applied Materials,
Inc. of Santa Clara, California, which was awarded $4.4 million to
develop ultra-high energy, low-cost lithium-ion batteries using a novel
manufacturing process. In the second, the Lab is working with Sion Power
Corp. of Tucson, Arizona, which received $5 million to develop
high-energy lithium-sulfur batteries for electric vehicles.
“The
government has given billions of dollars [in low-interest loans and
grants], venture capitalists are throwing money, and look at the number
of battery startups in the last few years. It’s gone from two or three
to dozens,” says Venkat Srinivasan, Berkeley Lab battery scientist and
Acting Group Leader of the Electrochemical Technologies Group. “The
great thing about the boom is there will be a lot of innovation.”
Battery Design: The Art of Trade-offs
Still,
no one expects a smooth road to 100 million EVs. Batteries are complex
electrochemical systems with some processes that even scientists don’t
completely understand. The wanted chemical reactions are accompanied by
unwanted side reactions that need to be controlled. The line between a
powerful, stable battery and a powerful, unstable battery is often a
thin one.
“On
the one hand, you’d like to drive your car for 300 miles on a single
charge. On the other hand, you have to realize you’re sitting on a
high-density energy source,” says scientist Robert Kostecki, who has
worked on batteries for 15 years and is also deputy director of Berkeley
Lab’s Environmental Energy Technologies Division. “The more energy you
pack in a small volume or small mass, the more hazardous behavior you
can expect.”
Making a battery is all about trade-offs. Srinivasan uses a spider chart (see diagram below; for technical version click here)
to show how present-day lithium-ion batteries compare to the DOE goals
for the FreedomCAR, a plug-in hybrid electric vehicle (PHEV) with a
range of 40 miles and a life of 15 years.
“It’s like a string,” says Srinivasan. “You pull on one end, you’re going to do something on the other end.”
For
example, to increase the energy density, typically the life of the
battery decreases, or the battery can be made safer, but then its energy
density will be lower.
At
Berkeley Lab, the focus is on lithium-ion batteries, which were first
commercialized in 1991 and are still considered the best near-term
option for transportation use. A “lithium-ion” battery, in fact, can
refer to any of a variety of different chemistries, and the Berkeley Lab
battery team is exploring a number of them. Which one will be the
eventual winner is not clear yet, and there may not be a single winner
because different applications have different requirements. While newer
alternatives such as lithium-sulfur and lithium-air hold great promise,
they will require technology breakthroughs before becoming a reality.
Part
of the motivation to jump-start battery innovation is to bring battery
manufacturing back to the United States. Production of lithium-ion
batteries, mostly for cell phones and other portable electronics, moved
to Asia, especially China, Japan and South Korea, nearly 20 years ago.
“China and Japan have spent 15 years gaining knowledge in the art of
making a battery. How do you beat that?” Srinivasan says. “You have to
think of a scientific way to approach this problem.”
The Key to Extending Life
In
principle, batteries are composed of a positively charged cathode, a
negatively charged anode, and an electrolyte solution that carries
charged ions between the two. When batteries fail, they can do so for
any number of reasons. Broadly, the causes fall into two
categories—mechanical degradation and chemical degradation.
“It’s
very hard to predict battery failure. We can’t simulate it,” Srinivasan
says. “Berkeley is trying to get to a battery simulator by getting to
the fundamentals of how batteries fail.
The
Berkeley team is also taking a fundamental scientific approach to the
chemical degradation by studying the protective layer that forms at the
interface between the electrode and electrolyte—the solid electrolyte
interface, or SEI. The SEI is one of the key components that enable
function of a Li-ion battery.
Stabilizing
the electrode/electrolyte interface has been pinpointed as critical to
extending the life of a battery. The SEI inhibits spontaneous
decomposition of the electrolyte—usually at the anode.
“Unfortunately,
we don’t fully understand how this layer forms and functions and what
it is made of,” Kostecki says. “It still escapes our best instrumental
techniques and experimental methodologies.”
While
batteries for cell phones and personal electronics are not expected to
operate much longer than two years, batteries for cars need to last at
least 10 if not 15 years.
“It’s
not a simple engineering extrapolation to extend life from two years to
15 years,” says Kostecki. “It’s a tremendous challenge. You have to
reduce the extent of the detrimental side effects in batteries by orders
of magnitude.”
The
SEI is a primary focus of research for Berkeley Lab battery scientists.
The team brings to the problem its strength in diagnostics and modeling
to detect and understand what is happening at the micro-, nano- and
molecular levels as the SEI forms, identify the critical processes, then
link those to the overall performance of the battery.
Cutting Costs
Another
requirement to getting a significant number of electric vehicles on the
road is cheaper batteries. Today’s lithium-ion batteries cost about
$1,000/kilowatt-hour. The DOE’s goal is to bring that down to $150/kWh,
which assumes a battery for an all-electric vehicle that can replace
what most people drive today, meaning a range of close to 300 miles.
“It’s
going to be very difficult to reach that goal,” Doeff acknowledges,
then adds, “It’s true we need to get the cost down, but I don’t know if
we need to get it down that far.”
Depending
on whether the battery is for an all-electric vehicle, a PHEV or an HEV
(hybrid electric vehicle, such as most Toyota Priuses on the road
today, which can go only a couple miles on its battery), the
requirements would be different. Doeff and Tom Richardson work mainly on
finding suitable materials for the cathode, one of the most expensive
parts of a battery, along with the separator and electrolyte solution.
The
most common cathode material in lithium-ion batteries is lithium cobalt
oxide. However, cobalt can be very expensive, and also tends to come
from countries that are not politically stable. “The long and short of
it is we have to get rid of cobalt to lower the prices,” Doeff says.
Some members of Berkeley Lab’s battery research team (Photo by Roy Kaltschmidt, Berkeley Lab Public Affairs) |
Other
cathode materials being looked at include lithium iron phosphate, which
is attractive because it delivers a good amount of power and iron is
inexpensive, but its energy density is inferior. It’s currently used in
power tools and is one of the top choices for hybrids and PHEVs where
power (acceleration) is of more concern than energy (range). The
challenge is to get more energy out of it.
Another
option is lithium manganese oxide spinel, advantageous because
manganese is inexpensive, although it too has lower energy density.
Doeff is also looking at titanium and aluminum as substitutes for
cobalt.
The
raw materials account for about 60 percent of a battery’s cost. The
remaining 40 percent goes to the manufacturing, a complex process that
can involve as many as 50 to 60 steps.
Reducing
manufacturing costs will require fundamental innovations in the way
batteries are made. It is an area ripe for change as the battery
manufacturing process has not evolved much since the voltaic pile was
invented 210 years ago.
“We
have materials scientists developing twenty-first century science. But
if you look at the way batteries are manufactured today, it’s not much
different from the original design that [Alessandro] Volta used in the
nineteenth century,” Kostecki says. “That discrepancy between the
innovation of state-of-the-art electrode materials and simplistic
manufacturing methodologies is one of the limiting factors for
lithium-ion batteries today. Manufacturing procedures currently are
largely based on trial and error. Consequently, electrode material
properties are seriously compromised by poor battery electrode design.”
For
example, graphite is the state-of-the-art material used in the anodes
of the vast majority of lithium-ion batteries. Lithium ions can travel
in graphite only between the graphene layers, but they cannot move
across this layered structure. Similarly, the electricity is only
conducted within the plane of the layers. However, graphitic carbons for
Li-ion battery applications have not been engineered to fully exploit
these properties.
“The
empirical way it’s done today is that battery companies contact the
graphite manufacturer, try all forms of graphite available on the
market, and then choose and optimize a selected few,” says Kostecki.
“Using a more rational approach to design graphite’s structure would
make electrodes perform better. I believe that materials scientists who
work on the next generation of electrode materials should work in unison
with engineers who can rationally design the battery electrodes and
cells, rather than separate these two functions, as they are now. It’s
an opportunity for Berkeley Lab to combine all of our resources and
approach this problem in a coordinated, holistic way.”
Making Sure Batteries Stay Safe
Another
important priority for Berkeley Lab battery researchers is safety,
which has been an issue in laptops and other consumer devices.
“Lithium
batteries do go up in flames occasionally,” says Richardson. “It’s
pretty rare, but once they burn, it’s hard to get them to stop. And
there are issues of toxicity.”
It
is precisely the advantages of lithium batteries—small in size and high
in energy—that make them potentially dangerous. Several factors could
cause a lithium battery to explode, including overcharging,
manufacturing defects and physical changes to the battery. Although the
odds of a single battery erupting are very small, an electric vehicle is
likely to have hundreds of them in a series, with current running
through each one. If the capacity of one cell is smaller than the
others, it will get overcharged, leading possibly to thermal runaway. To
deal with this, either the current could be diverted around the cell,
which would add weight and volume, or the array could be designed so
that it would just stop charging, which would limit the range.
Berkeley
Lab is developing an internal self-actuating overcharge protection that
would not significantly increase the weight or volume of the cell nor
the complexity of the manufacturing. To do that, Richardson and Guoying
Chen are looking at electroactive polymers, a class of polymers with
unique properties.
“The
polymer will get oxidized when the cell is being overcharged and go
from electrically insulating to conducting,” Chen explains. “So it
generates a short in the cell, between the anode and the cathode,
meaning no net current goes to the electrode and thus prevents the cell
from being overcharged.”
So
far, they have demonstrated that the concept works well with different
polymers as well as different cathodes and anodes. “And it’s reversible
too, so when you stop overcharging, the polymer goes back to being
resistive,” Richardson adds.
The work now is focused on finding a polymer and a configuration that will give optimal performance.
“How you put the polymer on the separator has a large effect, and also where you put it,” Chen says.
Wringing More Energy out of Lithium-Ion Batteries
On
the flip side of higher safety is higher energy density, which means
greater range for the vehicle. Within 10 years after lithium-ion
batteries were commercialized in the early 1990s, their energy density
doubled. Srinivasan believes it can double again within another decade.
There
are three ways to get higher energy density: increase the capacity,
increase the voltage or decrease the amount of inactive material in the
battery. The team at Berkeley is involved with all three aspects.
Materials research is being undertaken to find the next-generation
high-capacity cathode and anode materials, and new electrolytes that
allow the battery to operate at higher voltages without any detrimental
side reactions. For example, the Berkeley team has started research with
their partners in BATT to enable the use of a high voltage, stable
cathode that promises to increase the energy density compared to the
state of the art. In addition, the team has been pursuing avenues to
decrease the amount of inactive material in the battery while
maintaining the power capability and cycle life.
Assuming
success with innovative materials and processes, the timeline from the
lab to the marketplace is a long one for batteries. “People will tell
you it takes 10 years and $100 million to develop a battery system,”
says Doeff. “Even if we went into the lab next week and discovered the
next big thing that had everything we needed, it would still take 10
years to develop. These are seemingly simple devices, but there’s so
much we’re asking of them.”
BATT Program at Berkeley Lab
Venkat Srinivasan’s This Week in Batteries blog
Department of Energy’s Vehicle Technologies Program