‘Popcorn’ particle pathways promise better-performing lithium-ion batteries

FARID EL GABALY (8656) aligns an LFP Li-battery electrode sample for chemical characterization with X-ray photoelectron spectroscopy (XPS). The samples will then be thinly sliced for state-of-the-art synchrotron X-ray microscopy. (Photo by Jeff McMillan)

by Bruce Balfour

Sandia researchers have confirmed the particle-by-particle mechanism by which lithium ions move in and out of electrodes made of lithium iron phosphate (LiFePO4, or LFP), findings that could lead to better performance in lithium-ion batteries in electric vehicles, medical equipment, and aircraft.

The research is reported in an article titled “Intercalation Pathway in Many-Particle LiFePO4 Electrode Revealed by Nanoscale State-of-Charge Mapping” in the journal Nano Letters, 2013, 13 (3), pp 866-872. Authors include Sandia physicist Farid El Gabaly (8656) and William Chueh of Stanford University.

LFP, a natural mineral of the olivine family, is one of the newer materials being used in lithium-ion batteries and is known to be safer and longer-lasting than the lithium cobalt oxide (LiCoO2) compound used in smart phones, laptops, and other consumer electronics.

While LFP material is intriguing to researchers and battery manufacturers for those reasons, the process by which lithium ions move in and out of LFP as the battery stores and releases its energy is not well understood. This has proven to be a barrier to the material’s widespread adoption.

Cathode materials like LFP are critical in the search for higher-capacity, long-life, lithium-ion batteries for applications where batteries can’t be replaced as easily or as often as they are in consumer electronics. Larger applications where lithium cobalt oxide cells eventually could be replaced by LFP batteries include electric vehicles and aircraft.

‘Popcorn’-like particle movements seen

By observing complete battery cross-sections, the researchers have provided key insights on a controversy over the process that limits the battery charging and discharging rates.

Previous attempts to optimize the charging/discharging speed have included coating the particles to increase their electrical conductivity and reducing particle size to speed up their transformation, but have overlooked the initiation process that may well be the critical rate-limiting step in the way that lithium moves from a particle’s exterior to its interior.

By using X-ray microscopy to examine ultrathin slices of a commercial-grade battery, Sandia researchers found evidence that charging and discharging in LFP is limited by the initiation of phase transformation, or nucleation, and is unaffected by particle size.

The LFP electrode forms a mosaic of homogeneous particles that are in either a lithium-rich or lithium-poor state. The Sandia research confirms the particle-by-particle, or mosaic, pathway of phase transformations due to insertion of lithium ions into the cathode. The findings contradict previous assumptions.

“One propagation theory said that when all the particles were exposed to lithium, they would all start discharging slowly together in a concurrent phase transformation,” says Farid. “We’ve now seen that the process is more like popcorn. One particle is completely discharged, then the next, and they go one-by-one like popcorn, absorbing the lithium.”


Lithium ions move in and out of battery electrode materials as they are charged and discharged. When a rechargeable lithium-ion battery is charged, an external voltage source extracts lithium ions from the cathode (positive electrode) material, in a process known as “delithiation.” The lithium ions move through the electrolyte and are inserted (intercalated) in the anode (negative electrode) material, in a process known as “lithiation.” The same process happens in reverse when discharging energy from the battery.

“We observed that there were only two phases, where the particle either had lithium or it didn’t,” says Farid. “In many previous studies, researchers have focused on understanding the charging process inside one particle.”

Farid and his Sandia colleagues took a slice just a bit thicker than a human hair from a commercial-grade battery, just one layer of LFP particles, and mapped the locations of the lithium in about 450 particles when the battery was at different states of charge.

“Our discovery was made possible by mapping the lithium in a relatively large particle ensemble,” he says.

The researchers were able to build a commercial-grade coin-cell battery from raw materials using Sandia’s cell battery prototyping facility in New Mexico, which is the largest DOE facility equipped to manufacture small lots of lithium-ion cells. The battery was then charged, tested for normal behavior, and disassembled at Sandia/California using a new method of slicing layers that conserved the spatial arrangement from the cathode to the anode.

Characterizing the material

The Sandia researchers went to Lawrence Berkeley National Laboratory to characterize the materials with state-of-the-art scanning transmission X-ray microscopy at the Advanced Light Source (ALS), and then returned to Sandia’s California site for study by transmission electron microscopy (TEM).

“The X-ray spectroscopy from the ALS tells you what’s inside an individual particle, or where the lithium is, but it has low spatial resolution. We needed the electron microscopy of the same slice to tell us where all the particles were distributed across the entire layer of the battery,” says Chueh, a former Sandia Truman Fellow who is lead author of the journal article and an assistant professor and center fellow at the Precourt Institute of Energy at Stanford University.

Sandia’s research team and others presented their technical findings at the recent Materials Research Society Spring Meeting in San Francisco. As a result of that presentation, says Farid, other researchers are using the results to validate theoretical models. The team also may partner with industry, as one company has already indicated a strong interest in Sandia conducting similar studies on different, more complex battery materials.

The research team at Sandia has been funded internally, including support from the Sandia Truman Fellowship in National Security Science and Engineering, and by DOE’s Office of Science, which also supports the ALS.

-- Bruce Balfour

Back to top of page

Designing a more resilient nation, one building at a time

An aerial view of the damage Hurricane Ike inflicted upon Gilchrist, Texas. the last house standing on the waterfront at Gilchrist, Texas, in the wake of 2008’s Hurricane Ike demonstrates the value of engineering buildings to withstand the forces to which they are they are likely to be exposed. (FEMA photo)

by Neal Singer

Anyone who’s ever come home from vacation to find a home partly destroyed by a leaking roof, broken water line, or backed-up sewage knows the horror of drywall replacement, rotted rug ejections, mold tests, and other reconstruction measures that force life as we know it to a halt.

The difficulties are even larger when a commercial or government workplace is struck by disaster. Employees usually vacate the premises or wait for help. Little gets done.

Self-reliant buildings

But imagine a building resilient enough to allow people to continue their daily tasks even while an emergency is in progress. The self-reliant building might have its own small electrical generating system to maintain lights and computers during a power outage, employ mobile communications, maintain compartmentalized clusters of rooms to prevent water damage in one sector from affecting others, shatterproof windows to minimize hazards from imploding glass, and sufficient electronics to quickly pinpoint a trouble spot.

For obvious reasons, increased building resilience in the face of hurricanes, earthquakes, terrorism, or cyberattacks has been a major national security focus over the past decade.

Such resilient buildings not only would be less susceptible to damage and work interruption but could become community gathering places in times of general crisis, according  to a recently published Sandia paper, “Resilience certification for commercial buildings: a study of stakeholder perspectives,” published in Environment Systems and Decisions on March 13, 2013. 

But it won’t be easy to secure voluntary adoption by industry and construction companies if the wrong justifications are presented, says lead author Barbara Jennings (6924).

Expecting industry to act, for example, merely because “it’s the right thing to do” came out lowest (3 votes) in a questionnaire presented to 15 industry representatives.

The highest number of respondents were motivated by business reasons. These included increased revenue (10 votes), better competitive edge (9), and quicker, cheaper recoveries (9) from more efficiently handling a disruption. The upper middle ground was held by “decreased insurance premiums” (8) and “tax incentives” (7), while the lower middle ground included more problematic benefits: “Increased chance of receiving financing or lower finance rates” and “ability to charge higher lease rates due to increased attractiveness of the building to tenants” (both 5).

Five concepts

While the respondents were generally favorable to the resilience concept, they found it daunting to plow through the government forms and language necessary to apply for sizable tax credits to offset the increased building costs.

The paper proposes five concepts to make the concept of resilient buildings real to the construction, design, insurance, and building owner communities.

 Most imaginatively, the authors (who include Eric Vugrin and Deborah Belasich, both 6921) suggest that program sponsors collect stories and images that demonstrate resilience during alarming times and use these as proxy incidents to motivate others who had not experienced disasters themselves.

They also suggest government-based incentives, public-private partnerships, training and education programs, and simple, clear explanations of the federal governments’ multiple programs “to minimize confusion by describing the different role each plays.”

-- Neal Singer

Back to top of page

Detecting homemade explosives without also finding toothpaste

DETECTING HOMEMADE EXPLOSIVES — Chris Brotherton (6633) checks four tiny sensors in a test fixture, where he exposes the sensor to different environments and measures their response to see how they perform. Chris was principal investigator on a project aimed at detecting a common type of homemade explosive made with hydrogen peroxide. (Photo by Randy Montoya)

by Sue Major Holmes

Sandia researchers want airports, border checkpoints, and other security areas to be able to detect homemade explosives made with hydrogen peroxide without also singling out people whose toothpaste happens to contain peroxide.

That’s part of the challenge in developing a portable sensor to detect a common homemade explosive called a FOx (fuel/oxidizer) mixture, made by mixing hydrogen peroxide with fuels, says Chris Brotherton (6633). The detector must be able to spot hydrogen peroxide in concentrations that don’t also raise suspicions about common peroxide-containing products.

“Hydrogen peroxide explosives are a challenge because they are dangerous, but there are so many personal hygiene products that have hydrogen peroxide in them that the false positive rate is very high,” says Chris, principal investigator for an Early Career Laboratory Directed Research and Development (LDRD) project on chemiresponsive sensors to detect a common homemade explosive.

Hydrogen peroxide is found in everyday products ranging from soap, toothpaste, and hair color to laundry bleach, carpet cleaners, and stain removers.

The LDRD proved a sensor could identify relatively high concentrations of hydrogen peroxide and

differentiate that from a common interfering substance such as water, Chris says. The next step, he says, will be to work with an industrial partner to design an overall system that is faster and can be mass produced.

His work is built on field-structured chemiresistor technology developed at Sandia more than a decade ago by James Martin (1114) and Doug Read (1716). Chemiresistors are resistance-based sensors for volatile organic compounds. James and Doug, who have published several papers on their work, developed a significantly improved material that allows sensors’ response range and sensitivity to be tailored.

Finding the right polymer

Chris also faced the problem of coming up with a way to distinguish between hydrogen peroxide and water, which can exhibit similar behavior in chemiresistors. The key was choosing certain molecules in a polymer matrix, suggested by his technical mentor, polymer chemistry expert David Wheeler (1714). When exposed to peroxide, those molecules react in a different way than when exposed to water.

The idea is to engineer the polymer to be as similar to the target material as possible, relying on the undergraduate rule that like dissolves like. For example, David says, if the target is a substance that’s not very polar, you’d choose a polymer with nonpolar groups. If the target had a lot of polarity, like water does, you’d develop polymers that could hydrogen-bond with water.

The tiny sensor incorporates the polymer and chains of miniscule conductive metal beads. The polymer reacts when it’s exposed to the substance being analyzed.

“We tried to include specific molecules that would react with the peroxides,” Chris says.

Exposure to water also changes the polymer’s properties, but it returns to its previous state once the water is removed. Exposing the polymer to concentrated hydrogen peroxide, however, irreversibly changes it.

“So once you’ve done this to the polymer you’ve permanently changed it,” Chris says. “Instead of being a reusable sensor, it’s more of a disposable dosimeter.”

It’s also a detector that doesn’t react to toothpaste and other common peroxide products, he says.

Detector has other potential uses

Manager Paul Smith (6633) says the sensor has other potential uses, such as monitoring underground water, looking for plumes of contamination, or monitoring industrial processes.

Chris cautions that it’s not a silver bullet, but says the technology has shown good results.

“It has some challenges that have to be overcome, but we think it’s worth pursuing to the next level,” he says.

Researchers need to reduce the chemical reaction time so the sensor doesn’t take too long to be useful at a checkpoint, Chris says. The detector also must be incorporated into a larger unit that includes equipment to gather a sample for analysis.

The sensor doesn’t  need a significant amount of electronic processing or power supplies, Chris says, adding, “This technology would be easier to integrate into other detection technologies without impacting them too significantly.”

It wouldn’t have to be a large unit. Various detectors on the market today are about the size of a small handheld vacuum cleaner, Paul says.

Getting the air to the detector

The support equipment would suck up a sample of air and the detector would test it.

“You’d need to know where the fumes were coming from,” Chris says. “It’s not enough to open up the whole room and suck in all the air and say, ‘There are peroxides somewhere in here, watch out.’ What we’d like to do is go up and down luggage, or be next to some sort of industrial process so we know this is most likely the source and it’s above a level we care about.”

Although a detector package could target a single type of vapor, a manufacturer could add it to a unit to detect several substances. That way a checkpoint could have one sensing system rather than separate units for every material of concern, Chris suggests.

“Maybe it’s a suite of sensors to try to hedge our bets,” he says. “We’ve focused on a very specific application, but there’s no reason you couldn’t take this concept and use different polymers and look at multiple substances at the same time.”

-- Sue Major Holmes

Back to top of page

Download Lab News May 31, 2013 (PDF, 2MB)