Research

At BEST, Batteries Are Not Just a Black Box

A new Penn State lab is creating advanced lithium-ion batteries that will power a new generation of electric vehicles and store renewable energy.

Credit: ThinkstockAll Rights Reserved.

For decades, batteries have been just another commodity. One dies and we drop a new one in, whether it’s to power a cell phone or start an automobile.

Even research scientists treat batteries as a kind of black box, says Chao-Yang Wang, Distinguished Professor of Mechanical Engineering at Penn State. “They look at the ratings and plug them in, but they never open them up. They don’t know whether an application is limited by the battery’s anode, cathode, or something else.”

All that may be about to change.

Batteries—especially rechargeable lithium-ion (Li-ion) batteries—are undergoing a renaissance. They drive nearly all portable electric devices, and are critical in hybrid electric and all-electric vehicles. They promise to store solar and wind power for use when the sky is cloudy or the air is still. They could power an economy that is greener, more sustainable, and less dependent on imported oil.

These promising techologies rely on large Li-ion batteries, which differ radically from their smaller cousins. They may involve new chemistries, new structures, and complicated control systems. Penn State’s new, 10,000-square-foot Battery and Energy Storage Technology (BEST) Center has become a focal point for development of vehicle and grid-sized Li-ion batteries.

smiling man in labcoat with lab in backgroundM. Scott Johnson

Chao-Yang Wang, Distinguished Professor of Mechanical Engineering

Located in the former Materials Research Laboratory, BEST is co-directed by Wang and Chris Rahn, a professor of mechanical engineering. It incorporates six other principal researchers and their students, and draws on faculty from the College of Engineering, the Eberly College of Science, and the College of Earth and Mineral Sciences.

“There is definitely expertise around the country in battery devices, materials, and systems,” Rahn notes. “As far as I know, though, we are one of the first groups to house that expertise together in one building.”

Wang sees something else that makes BEST unique. “We are the first U.S. university lab capable of fabricating a full-sized battery that would fit inside an electric or hybrid electric car,” he says. “Lots of laboratories can make small, coin-sized batteries. But very few engineers trained in the United States know how to manufacture large lithium-ion batteries.

“We’re part of a government initiative to establish an advanced battery manufacturing industry in the United States,” Wang states. “Until now, there was literally no large lithium-ion battery industry in this country. Everything was done in Asia. We are going to train next-generation engineers who will compete in this area.”

Making Batteries

U.S. industry has lots of experience with conventional lead-acid vehicle batteries, Wang adds, but manufacturing Li-ion poses new challenges. Lithium is among the most reactive of all metals. Expose it to moisture and air and it burns. This calls for special handling, especially when manufacturing and assembling large objects.

Then there’s the chemistry. Lead-acid batteries like those in conventional automobiles are simple. They consist of two lead electrodes, an anode and cathode, separated by an insulator and immersed in a sulfuric acid electrolyte. Discharging the battery breaks the sulfuric acid down into positively and negatively charged ions that provide electrical power.

microscope view of mesoporous metal oxideCourtesy Donghai Wang

Mesoporous metal oxide, a material with potential for next-generation energy storage.

Lithium-ion batteries are more complex. Like lead, lithium is used in both electrodes. Unlike lead, it is always combined with other materials. Popular lithium cathodes include lithium cobalt oxide, lithium iron phosphate, and lithium manganese oxide. Lithium anodes include lithium carbon, lithium titanate, and lithium silicate. Electrolytes offer an even broader range of chemistries.

BEST’s Battery Manufacturing Laboratory handles all these materials. For this reason, staffing the laboratory presented an unusual challenge. After all, few academics in America had any experience with Li-ion battery manufacture.

Wang and his associates have helped close that gap. “We know from our research how lithium-ion batteries are made. We’ve traveled around the world and visited many factories. We recruited faculty with experience, and we supplement their expertise with post-doctoral students who have practical experience in Korea, Japan, China,” Wang explains.

Many large companies, from automakers (Ford, GM, Honda, Hyundai, Toyota) and battery manufacturers (Eveready, Panasonic) to chemical companies (Cabot, Gore, Conoco) and national laboratories (Argonne, Lawrence Berkeley, NIST, National Energy Technology Laboratory), collaborate with BEST to gain access to that expertise. Often, that expertise comes in the form of new graduates. “What the University is doing is training a very large number of engineers,” Wang says. “If we are going to have a battery industry in the United States, we must have a pipeline of engineers with skills in research and manufacturing.”

Systems

Like many scientific enterprises, BEST traces its beginnings to a seemingly simple question: How do we control large electric vehicle battery systems?

At the time, Wang was just starting to look at batteries. He came to Penn State in 1997 after creating a reputation at the University of Hawaii for his work with fuel cells, which produce clean energy from hydrogen. Yet fuel cells had limitations. Not only were their platinum catalysts costly, but widespread use would require an expensive national infrastructure to make and distribute hydrogen.

Chris Rahn

Chris Rahn, Professor of Mechanical Engineering

By the early 2000s, batteries had caught Wang’s attention. Innovations abounded as researchers sought more powerful, long-lasting batteries for portable electronics. In 2001, Wang took on a graduate student, Kandler Smith. He wanted to understand how to control large battery systems that could be used in hybrid and all-electric vehicles. “This was back before the Prius had made a big splash,” Rahn recalls. “Nobody was thinking about how to build complicated battery systems that you could charge and discharge thousands of times.”

When Rahn himself arrived at Penn State, a few years after Wang, it was a homecoming. His parents had met while graduate students at the University. One of their earliest videos shows him running circles around his father’s legs during graduation. Rahn specialized in modeling electronic control systems for mechanical systems. Soon after his arrival, Smith realized that such models would play a crucial role in controlling large battery systems and asked Rahn to be his co-advisor.

Rahn, who knew nothing about batteries, was not enthused. “Fuel cells were very hot at that time, and I was hoping we’d start working on them. The battery stuff didn’t seem very exciting.” Still, the problem intrigued him. A single battery is relatively simple, only electrodes, insulators, and electrolyte. Large systems consist of hundreds of these smaller cells. Each cell charges and discharges at similar but varying rates. Some fail sooner, others later. The researchers needed a way to control the charge and discharge of all those cells, and understand how they would perform under different conditions.

If this sounds simple, consider something most automobile drivers take for granted: the fuel gauge. In a conventional vehicle, it links to a float in the gas tank. As the car burns fuel, the float falls and the needle on the dashboard moves towards empty.

Battery-powered devices incorporate something that resembles a fuel gauge, too. On cellphones, it shows up as a power bar indicator. Its technical name is the state-of-charge estimator. But unlike a mechanical fuel indicator, which directly reads the amount of fuel in the tank, it cannot peer into the battery. Instead, true to its name, it estimates the amount of power left. It does this by measuring changes in battery voltage. As the battery uses up its charge, its voltage declines.

In cellphones, where batteries discharge slowly over a long period of time, this decline is predictable. But electric vehicle batteries discharge when they accelerate and recharge when they brake, a cycle that occurs dozens or even hundreds of times during a trip. As anyone who has ever charged a cellphone for a few minutes knows, partial charges make batteries look stronger than they are—for a few moments, anyway. Electric vehicles receive many partial charges, and it’s easy for a state-of-charge estimator to make mistakes.

“A hybrid charges and discharges minute by minute,” Rahn says. “Having that information for the hundreds of cells that make up a battery pack is very critical. If you don’t have a good state of charge estimator, you could wind up stranded.”

Models

Clearly, a reliable system could not rely merely on measuring voltage. Instead, the researchers needed to understand the battery chemistry well enough to predict how each cell would react to ongoing partial charge-recharge cycles. Then they could represent these behaviors with a family of interlinked equations, or model, which would then calculate the state of charge.

Smith and Wang turned to Rahn because he specialized in distributed parameter systems, where variables are unevenly distributed in space. Batteries are this type of system. The ions that carry electrical charges are distributed between the anode and cathode. “If you can figure out how the concentration changes when you charge or discharge, then you know how the battery will perform,” Rahn says.

A model could make charge/discharge decisions every fraction of a second. “If I’m braking, it can look and see whether I should send the charge back into the battery,” Rahn explains. “If the batteries are already too charged or the rate at which the energy is coming back is too high, I have to say, 'Sorry, the battery can’t take it.’” The researchers could also use the model to test new battery designs and to optimize size and recharging times without having to build hundreds of prototypes. The model is an essential tool in designing the most economical, powerful battery, which is by far the expensive part of an electric vehicle.

Such models were not easy to build, however, even with Wang’s detailed knowledge of fundamental battery chemistry. To create a model of a single cell, the researchers divided the space between electrodes into very fine slices, then calculated how different conditions affected each slice, and how each slice’s results influenced the slices around it. The models were so complex, they required a small supercomputer to run.

Vehicles, however, have only a small processor to spare. So, once they had accurate models, Rahn, Smith, and Wang used mathematical tools to approximate their results using simpler versions. They continue to work on improving performance and extending models to new Li-ion chemistries.

locomotiveMichael Bezilla

BEST Center engineers helped optimize the design for Norfolk Southern’s No. 999, the first all-electric, battery-powered locomotive in the United States. Read more...

New Research

BEST is pressing ahead on other fronts as well. Its work ranges from materials and structures to integrated systems and the technology needed to manufacture them.

Wang, for example, focuses on structures that can handle the challenges presented by larger batteries. Sometimes this involves testing new materials that are rich in lithium, inexpensive to make, and thermally stable. The latter is especially important in large batteries. Even small Li-ion batteries tend to warm when recharging. “Now imagine that you took 1,000 laptop batteries and put them under the hood of a car. If they all started heating up to 100 or 120 Celsius, they could cause a fire or an explosion,” Wang says.

Structures determine how efficiently a battery uses its active materials. In small batteries, this is rarely an issue: Designers simply place thin electrodes close to one another and lithium ions flow easily between them. Large batteries, however, require different strategies. They have thicker electrodes separated by greater distances. Lithium ions must travel further between electrodes. The electrodes themselves are made of powder in a polymer binder, so the ions can penetrate the surface and reach the active material.

Managing distance and porosity in large Li-ion batteries is a difficult juggling act. “It is why researchers get high efficiencies when they demonstrate small samples, but when they try to build a practical battery, the numbers are not as good,” Wang says.

As part of his search for a better balance of properties, Wang optimizes structures for particular applications. A hybrid, for example, needs a burst of power to accelerate before its gasoline engine kicks in. An all-electric vehicle, on the other hand, needs to provide power mile after mile. Controlling electrode thickness—thinner electrodes produce greater bursts—provides one way to control this.

Another Wang project is the water battery, which stores renewable energy. It acts like a combination of storage battery and fuel cell. Electricity from wind turbines or solar cells splits the battery’s water into hydrogen and oxygen. It stores the hydrogen, and burns it like a fuel cell when it needs to make electricity.

Unlike conventional fuel cells, a water battery does not require expensive platinum catalysts. “Without platinum, we can afford to build them bigger,” Wang says. “They discharge slowly, but we offer 20 times more energy density than lead-acid batteries and 10 times longer life.” He is currently building a 10-kilowatt prototype, equivalent to the power in 10 car batteries.

Other BEST researchers are pursuing other aspects of advanced batteries. Michael Hickner, for example, works on the polymeric binders that hold together the active material powders of the electrodes.

Creating a binder is more complicated than just adding glue. Hickner wants to use as little binder as possible, so more of the structure consists of active powders. Binders must also accommodate changes in volume as the electrodes swell and shrink during charge and discharge. Hickner uses specialized imagery to determine how binders degrade during use, then redesigns their polymers to address their weak points.

Another researcher, Donghai Wang, is investigating next-generation batteries based on a sulfur cathode. They offer tantalizing possibilities: They run cool and will not burst into flames, and they have ten times the capacity of existing cathodes. Sulfur not only costs far less than lithium, but it uses greener chemistry and permits more stable electrolytes and interfaces.

Wang’s research is still in the science stage. His small sulfur cathodes fail after several hundred cycles. Before building larger batteries, Wang will have to raise that to the thousands of cycles necessary for electric vehicles.

BEST’s research is pointing the way toward advanced batteries and the innovative applications they make possible. Equally important, BEST is training researchers who will play an important role in creating cleaner, greener, and more sustainable vehicles and renewable power.

Chao-Yang Wang, Ph.D., is Distinguished Professor of Mechanical Engineering, and Christopher Rahn, Ph.D., is professor of mechanical engineering, both in the College of Engineering. Wang and Rahn are co-directors of the Battery and Energy Storage Technology (BEST) Center at Penn State.

Last Updated January 25, 2012