In 1747, Benjamin Franklin received a glass jar coated in metal foil from a colleague in London. He called it “this miraculous bottle,” and set about figuring out what it could do.

To many, the Leyden jar was a strange curiosity, a toy for sparking electric shocks, making monks jump into the air, or sending jolts through a chain of guardsmen for the amusement of the French king.

Franklin saw it differently. For the polymath and Founding Father, the jar was a way to capture electricity, store it and put it to work — something that had never been done. Later that year, he connected 11 jars together. In the process, he coined a new term: “We made what we call’d an Electrical Battery,” he wrote.

Franklin’s mind was famously restless. Nearly three decades later, in 1776, he applied that same spirit of experimentation to a different question: the issue of how an independent America might be governed.

Could power itself be structured and wielded according to reason and evidence? For Franklin, science and politics were expressions of the same instinct: to question traditional authority and insist on testing ideas for yourself.

That instinct drives scientific discovery and — thanks in part to Franklin — is woven into the DNA of the American experiment. It is thriving two and a half centuries later at USC Dornsife, where scientists revisit questions others have dismissed or overlooked.

Among these scientists is mathematician Aaron Lauda, who discovered his own “miraculous bottle” in a set of equations earlier researchers had discarded.

Computing on the Edge”

Quantum computing is one of the great frontiers of modern science. By harnessing subatomic particles to follow many computational paths at once, quantum computers promise to process information in fundamentally new ways. In theory, their capabilities are immense, from simulating complex molecules to breaking certain forms of encryption and solving problems beyond the reach of classical machines.

Quantum computers work by harnessing qubits — quantum bits that, unlike the binary zeros and ones of a traditional computer, can exist in multiple states simultaneously. But qubits are extraordinarily fragile. The quantum states they rely on can be disrupted by stray vibrations, which is why no one has yet built a system capable of realizing their full potential.

“All the things that we want to do — simulate new drugs, break encryption — need thousands and thousands of qubits,” says Lauda, divisional dean for math and physical sciences. “And we’re very far from having that scale.”

One promising approach would protect quantum information by encoding it in the pattern of a path that remains stable even as the environment shifts around it. For years, however, that approach seemed to hit a wall. The particles best suited to the task could only perform a limited set of operations, a ceiling mathematicians treated as a fact of nature. Lauda, who is also professor of mathematics and physics and astronomy, realized that it was instead an assumption so old that nobody had properly questioned it. Researchers had previously identified a vast class of things the particles could do but dismissed most of them as useless. For decades, no one looked again. But Lauda did.

What he found was that refining how those capabilities are measured transforms apparently useless characteristics into powerful ones, massively expanding what the system can do. The work points toward a new architecture for quantum computing. “It requires us to push the paradigm,” he says, “of how we think about these approaches.”

Mapping Silent Collisions

While Lauda works at the level of subatomic particles, Kris Pardo, a cosmologist, is searching at the opposite extreme. Pardo explores mergers of black holes billions of times the mass of the sun, events so violent they warp spacetime itself.

Scientists know such mergers must happen but have never detected one. The gravitational waves they produce oscillate at frequencies that existing detectors do not capture.

But Pardo, assistant professor of physics and astronomy, had an ingenious idea. What if data gathered to find planets could also reveal signals no one had thought to look for?

The Kepler Space Telescope, launched in 2009 to search for planets, had amassed an enormous trove of data by the time its mission ended in 2018. Pardo realized that buried deep in that archive might be subtle signals of gravitational waves, which can cause stars to shift their apparent positions in a slow celestial ripple.

“The data was meant for finding planets,” he says. “But it also contains precise measurements of stellar positions.”

Pardo and his group are currently working their way through that vast dataset. “This would be the only way to observe supermassive black hole mergers — collisions of black holes 10 million to a billion times the mass of the sun,” he says. Even if they ultimately prove elusive, the effort will still advance the science.

“We’re getting close,” Pardo says. “But we’re not there yet.”

Engineering a Cleaner Future

If Pardo is scouring the universe’s distant past, chemist and Distinguished Professor Professor Surya Prakash is focused on the near future, and on a green energy solution hiding in plain sight.

“Carbon dioxide is seen as a problem,” says Prakash, the George A. and Judith A. Olah Nobel Laureate Chair in Hydrocarbon Chemistry and professor of chemistry and chemical engineering and materials science. “But it can also be a solution, if we capture and use it.”

For nearly 40 years, the director of USC Dornsife’s  Loker Hydrocarbon Research Institute has pursued an unusual path in the search for clean energy. “CO₂ is a miracle molecule,” Prakash tells his students. “It’s how photosynthesis works. That’s the foundation of life on Earth.”

If carbon dioxide is fundamental to life, Prakash reasons, it can also be part of the solution. By combining captured CO₂ with hydrogen produced from renewable electricity, it is possible to make methanol, a versatile, energy-dense liquid used to power engines and manufacture plastics and pharmaceuticals. A plant in Iceland, inspired by Prakash’s work, is already using his method to produce methanol commercially.

“One-carbon problem,” Prakash says, “one-carbon solution.”

But turning that idea into a global reality depends on making the chemistry fast and efficient enough to work at scale. That’s the challenge Smaranda Marinescu, associate professor of chemistry, is tackling. Her lab is designing catalysts that drive the conversion, at molecular scale, of CO₂ into usable fuels.

“We take inspiration from nature,” Marinescu says of her work, which draws on the way living systems convert energy. Her team has developed a catalyst that converts CO₂ to methanol with 42% efficiency. “We’re currently improving on that, and because the catalyst is electrochemical, it’s possible to drive the whole process with renewable electricity,” she says.

Marinescu and Prakash’s work on clean fuels builds on an enduring legacy at USC Dornsife. In 1994, George Olah — Prakash’s longtime mentor and collaborator — was awarded the Nobel Prize in Chemistry, the university’s first. Olah, who died in 2017, often argued that methanol was the successor to oil.

But Olah was not the only Nobel laureate in chemistry at USC Dornsife. Another chemist’s route to recognition was far less conventional.

Rejection Fuels Discovery

Denied tenure in Israel for work that would later help him earn the 2013 Nobel Prize in Chemistry, Arieh Warshel became a pioneer in computational models for complex chemical systems. His work revealed how enzymes function at the molecular level and helped lay the foundations for computer-aided drug design.

The Distinguished Professor of Chemistry has never had much patience for orthodoxy. “A lot of my papers were rejected before they were accepted,” he says.

Now in his ninth decade, Warshel leads USC’s Warshel Center for Multiscale Simulations, where his graduate students enjoy the freedom to study everything from bird navigation to drug resistance to enzyme design.

After arriving in the United States half a century ago to join USC Dornsife, he is still at his desk running simulations and is now turning to artificial intelligence to tackle problems that resisted his earlier methods.

Asked why, at 85, he has no intention of stopping, Warshel doesn’t hesitate. Like Franklin 250 years before him, he is still asking what happens when you refuse to accept limits. “What are the other options?”