The Battery Revolution

The Battery Revolution

The Battery Revolution

Assistant Professor Yifei Mo works in Deepthought2, UMD's high-performance computing cluster. Photo: John T. Consoli
Assistant Professor Yifei Mo works in Deepthought2, UMD's high-performance computing cluster. Photo: John T. Consoli

Fires in cell phones, laptops, and even a jumbo jet might have one thing in common: a liquid-electrolyte lithium-ion (Li-ion) battery.

Flammability—coupled with Li-Ion batteries’ limited performance and consumers’ insatiable demand for long-lived, portable, and lightweight electronics—drives researchers around the world to develop an alternative power source. Aided by artificial intelligence (AI) methods such as machine learning (ML), University of Maryland engineers are among the leaders in the race to build a better battery—the ultimate battery.

Explains Yifei Mo, assistant professor of materials science and engineering, “What we’re after is ultimate safety, energy density, life cycle, and charging time.”

In Li-ion batteries, positively charged particles called ions move through a liquid electrolyte between positive and negative electrodes. However, the liquid electrolyte limits power density when ions are not transported rapidly, causes slow recharge time, and risks bursting into flame. A new electrolyte base is needed to create batteries capable of moving more ions at a time, called fast-ion transport, and capable of discharging a higher electrical load.

ML can help find it.

A computational specialist, Mo and his collaborators use supercomputers aided by ML to model materials, atom by atom. This enables them to predict how the materials will conduct ion flow and create electric current, thereby speeding discovery of highly promising, novel solid-state battery materials.

“Fast-ion diffusion is very rare in solids,” says Mo, who also holds an appointment with the Maryland Energy Innovation Institute. “Only a handful of materials can achieve this at room temperature. Ion transport is determined by many factors of the materials, and many are still not fully understood.”

Mo’s lab is among the first to employ ML techniques to accelerate materials discovery. Through the ML training on existing materials data, promising new solid-state battery materials that exhibit the required property of fast-ion transport may be found.

Mo and his research team have filed multiple patents on potential new materials for solid-state batteries and are working with industry partners to develop them.

Their next steps include testing novel materials and bringing them to consumers. “We want to commercialize new battery technology that revolutionizes our battery-powered life, everything from cell phones to cars,” he says.

Related Articles:
Advancing Healthcare through Robotics and Machine Learning
Algorithms and Autonomous Discovery
Machine Learning's Translational Medicine
The Buddy System: Human-Computer Teams
Measuring Change in the Atmosphere
Manocha Receives 2022 Verisk AI Faculty Research Award
Clark School Faculty Receive CAREER Awards
Fuge Receives NSF CAREER Award
AlphaGo family of AI programs grew from AMS simulation-based algorithms developed at UMD
Measuring Snow From Space

July 10, 2019


Prev   Next
“What we’re after is ultimate safety, energy density, life cycle, and charging time.”

Yifei Mo



Current Headlines

UMD Joins $50M Sodium-Ion Battery Innovation Partnership

Celebrating Five Years of Innovation at CEEE’s Daikin Lab

Project Embraces Tribal History With Modern Technology

Former Chair of Materials Science and Engineering To Retire from the University

Sophomore in Chemical and Biomolecular Engineering Heads to NCAA Cross Country Championship

Eminent Scholar in Metallurgy To Join Clark School as Distinguished Chair

UMD Joins Sodium-Ion Battery Alliance for Renewable Grid Energy Storage

Biocomputational Engineering Program at UMD Earns ABET Accreditation

News Resources

Return to Newsroom

Search News

Archived News

Events Resources

Events Calendar