Close Mobile Menu

Theorist’s Tools: Marvin Cohen Uses Quantum Mechanics and Computers to Conjure Future

June 18, 2015
by Sabin Russell

Since the early 1960s, UC Berkeley theoretical physicist Marvin Cohen has been predicting the shape of things to come.

His name will be forever linked to the deep and complex physics behind semiconductors—the active ingredients of computer chips, flat-panel TVs, and solar panels. Cohen was a pioneer in the use of computers to predict the electronic traits of silicon under different conditions. His modeling of graphene, a honeycomb sheet of carbon as thin as a single atom, may yet help yield a successor to silicon as the fundamental driver of computer technology. And his theory-driven forecasts still guide the search for superconductors—exotic metals that might someday transmit electric power effortlessly across continents.

The Nobel committee has yet to call, but today there is growing recognition of Cohen’s contributions, and he is collecting accolades, including the 2001 National Medal of Science and the 2011 Dickson Prize in Science from Carnegie Mellon University. Last fall, he was named the recipient of the 2014 Von Hippel Award, the highest honor of the Materials Research Society, a global organization of scientists from multiple fields who dream up and develop the remarkable arrangements of matter that will likely shape our future.

In March, Cohen not only celebrated his 80th birthday, he presided over the opening of an airy and light-filled room—named in his honor—just down the corridor from his own top-floor office in Birge Hall. The new room is equipped with an espresso machine, comfortable chairs, and translucent glass panels of dry-erase boards—all of it designed to lure his fellow thinkers back to the old-school practice of talking physics with each other, face-to-face. “We can write on these walls, and sit at these tables, and drink coffee and talk,” says Cohen. “And hopefully great ideas will be generated there.”

Materials physics is a collaborative enterprise, and Cohen believes that something is lost when researchers compose an email for a colleague in the next room, rather than simply walking over to ask the question or talk about a fresh idea in person. Back in the day, he says, grad students would schlep up the hill to the computers housed at Lawrence Berkeley National Laboratory, carrying their programs and data in stacks of IBM cards. In the hours it took to process those cards, the physicists would talk to one another, and out would pop fresh ideas. It is the very imprecision of casual conversation, Cohen says, that pulls thinking out of ruts and invites creativity. “When there’s play in the system, and all the cogs are not exactly lined up, sometimes it is better,” he adds.

The new salon for physicists harkens back to the “afternoon teas” at Bell Labs in Murray Hill, New Jersey, where Cohen spent a year as a postdoc. “I was very impressed that at 4 o’clock we’d go for tea and cookies, and I would have a chance to talk to a lot of the senior people. You’d hear the latest gossip, the latest this and that,” he recalls.

Bell Labs actually designed interactions into the architecture of its research centers, where fundamental innovations including the transistor, the laser, solar cells, communication satellites, and cellular networks were born. Theorists would encounter experimentalists along the immensely long corridors connecting offices, where doors were meant to stay open. Former Berkeley Chancellor Robert Birgeneau, who was a physicist at Murray Hill from 1968 to 1975, fondly remembers those tea sessions.

“The level of discussion was extraordinary,” says Birgeneau, who returned to laboratory research in 2013 at the end of his eight-year run as head of the University. “Various physicists would wander in and out of these teas. They included Phil Anderson, Steve Chu, Doug Osheroff, Horst Störmer, Dan Tsui, and Bob Laughlin, all of whom won the Nobel Prize in Physics.”

Although Cohen retired from university teaching in 2010, he returns to his office daily, and continues to travel the globe working with former graduate students, postdocs, and people trained by his own students. “Marvin is a marvelous teacher who has educated about two generations of the best solid-state theorists, not only in this country, but around the world,” says Birgeneau.

Guests at the dedication of the new physics hangout included former students from as far away as Brazil, China, and South Korea. Among them was Renata Wentzcovitch, who was a grad student from São Paulo when Cohen began schooling her in the predictive power of his mathematical equations. Today she is a materials physicist at the University of Minnesota. By calculating how minerals change under extreme pressures and temperatures, she is building an understanding of the makeup and dynamism of the Earth’s mantle. She has become a leading expert on magnesium-silicate perovskite, the most abundant material within our planet.

“This mineral alone may account for 50 percent of the Earth’s mass,” she notes. Her calculations have even been applied to model the innards of newly discovered exoplanets—giant, rocky, super-Earths that orbit distant stars.

Birge Hall, located just east of Sather Tower, has long been home to the Condensed Matter Physics and Materials Science group, the largest contingent of Berkeley’s highly acclaimed physics department. Cohen, like nearly all the top physicists there, also holds a joint appointment at Lawrence Berkeley National Laboratory, the U.S. Department of Energy lab on the hill overlooking the Berkeley campus.

Cohen’s lair is a book-lined office with a window on the Campanile and the requisite portrait of Einstein on one wall. A cheerful raconteur who speaks with the relaxed demeanor of a patient uncle, Cohen—like the Father of Relativity himself—mixes a love of mathematics with a passion for music. Einstein had his violin, Cohen his clarinet. During his high school years, Cohen played improvisational jazz on an alto saxophone; now he masters the classics with the clarinet. He tries to play for an hour every day.

“I see a lot of people in my field who play instruments,” he says. “Unlike with jazz, when I am playing Brahms or Mozart I’ve got to get everything as close to what Brahms or Mozart wanted. I mean: They lay down the law.”

Cohen was born in Montréal, but his parents moved to the United States, and he grew up in San Francisco’s Richmond District. There he played in high school jazz bands that featured a schoolmate he had known since Roosevelt Junior High. His name was Johnny Mathis—that Johnny Mathis, the honey-throated crooner of Chances Are and Misty who became one of the most successful pop singers of his time. “I once had some great advice for Johnny Mathis,” Cohen recalls. “I told him never to sing popular music; only jazz…. I use that with my students sometimes, as an example that shows I don’t always provide the best advice.”

Cohen traces his love of physics to an insight he gained as a 7-year-old tossing a ball with friends in Montréal. “A high school boy came by and told us that he had learned in school that he could tell exactly where the ball was in the air, at any point, by using a math equation,” he recalls. “I was amazed by that and, it turns out, I’m still amazed…. The idea that I could use math and physics to explain and even predict what happens in nature is the driving force of my professional life.”

Pic­ture a phys­i­cist filling a lec­ture-hall black­board with equa­tion after equa­tion. How many such black­boards go in­to his com­pu­ter­ized por­traits of a new struc­ture of car­bon? “Hun­dreds.”

Even as he begins his ninth decade, Cohen remains dedicated to sharing that fascination. He still visits local high schools, counsels community college science students, and loves to challenge his three teenage grandchildren with math problems. (Their mother, Susie Cohen Crumpler, is chief marketing officer for the Cal Alumni Association, publisher of California.)

Cohen still keeps in close touch with some of his high school friends, who say it was not obvious then that he was destined to become the acclaimed physicist he is today. “He definitely did not come across as studious,” says Lee Battat, who remembers Cohen as a fun-loving, popular teenager. “He was never a snob, never self-important.”

Lee is married to Cohen’s high school chum, Frank Battat. Both young men became physics majors at Berkeley. They were fraternity brothers and have remained friends to this day. Cohen also remains close with Steve Krieger, another San Francisco pal and fellow physics major at Berkeley. Krieger and his wife Arlene became the founding benefactors in financing the renovations of Cohen’s new Birge Hall room, officially called the Marvin L. Cohen Interaction Area.

“Interaction” is a word that goes to the heart of what materials scientists like Cohen do. Their day-to-day preoccupation is the interaction of particles, which attract or repel each other in behavior governed by the bizarre and mind-boggling rulebook of quantum mechanics. In the spooky world of quantum physics, matter behaves as both a wave and a particle, and interactions within solid materials are described in measures of electron volts, infinitesimal units of energy. Cohen’s computer programs deploy the exquisite equations of quantum theory to explain the form and function of known materials, and are used to forecast the performance of newly envisioned ones. “Quantum mechanics,” he says, “fed my family.”

Make no mistake: Cohen’s computer programs are extraordinarily complex. Picture a physicist filling a lecture-hall blackboard with equation after equation. How many such blackboards go into his computerized portraits of a new structure of carbon? “Hundreds,” answers Cohen. “The advantage of the computer is that you can do more than you could possibly do with a pencil and paper. But you have to do the pencil and paper work first, to set the problem up. Only then can the computer give you numbers.”

Vincent Crespi, a former student of Cohen’s who is now a physics professor at Penn State, says that the computational tools developed by Cohen shortened the distance between theoretical ideas and practical applications. The interactions of trillions of electrons inside a solid are so difficult to track mathematically that they can wear out even the fastest supercomputers. Using much slower and smaller machines that were state of the art decades ago, Cohen figured out ways to make end-runs around the complexities. The term “shortcut” would give his work short shrift, but Cohen’s masterstroke—known as the “empirical pseudopotential method”—brilliantly simplified otherwise impossibly tangled problems. As a result, says Crespi, “it allowed theorists to make predictions about materials, not just seat-of-the-pants guesses.”

If sci­ent­ists can dis­cov­er how to make graphene in­teg­rated cir­cuits eco­nom­ic­ally, it may pre­vent the mi­cro­elec­tron­ics in­dustry from run­ning out of mir­acles in a few short years.

Many of his physics colleagues cite Cohen’s computer programs as crucial to the development of the semiconductor industry. “If you want to study silicon, you can’t just dabble around blindly in the lab,” says Berkeley experimental physicist Alex Zettl, a longtime collaborator with Cohen. “You really need to understand the electronic structure of silicon. And Marvin’s theories explained silicon.”

Zettl graduated with a B.A. from Cal in 1978, but it was only after he returned to Berkeley as a young faculty member five years later that a long and fruitful partnership with Cohen began. Cohen is a theorist, and Zettl an experimentalist. Together, they have coauthored scores of papers. “Marvin has a way of paring complicated problems down to the bare essentials,” says Zettl. “He makes predictions about materials, and I try to synthesize them in the laboratory. If the properties turn out to be unusual, he goes back and tries to explain them. It’s sort of a cycle.”

This synergy between theorist and experimentalist is the secret sauce of basic research, and it drives the discovery of new materials such as graphene, a one-atom-thick grid of carbon that is far stronger than steel. Experiments have shown that, depending on how it is structured—flat as a sheet, scored like wavy corduroy, cut into ribbons, or rolled in various ways into tubes—the graphene “nanomaterial” has a wide variety of electronic properties that can mimic or surpass those of silicon. Meanwhile, ever-shrinking silicon-based transistors are nearing absolute size limits, threatening to end the run of what is known as Moore’s Law—Intel cofounder Gordon Moore’s uncannily accurate forecast that the number of transistors on integrated circuits would double every two years. Devices made of graphene, however, can work at even smaller geometries, and if scientists can discover how to make graphene integrated circuits economically, it may prevent the microelectronics industry from running out of miracles in a few short years.

In 1994, Cohen was flying home from a scientific conference when he began to think about other materials that could be rolled into “nanotubes” the way graphene can. With pencil and paper in hand, he began to sketch out calculations for a nanotube made of boron and nitrogen atoms in place of carbon. Back in Berkeley, he had a graduate student work out the details. The calculations predicted fascinating properties. A year later, Zettl made microscopic amounts of boron-nitride nanotubes in his lab and found that the “white graphene” performed just as predicted. “His theory was right on the money,” Zettl recalls.

“Today, the gen­er­a­tion, trans­port, and stor­age of en­ergy is very im­port­ant. Every five years or so, someone will dis­cov­er something sur­pris­ing, and that’s what makes this field so ex­cit­ing.”

White-graphene fiber made of these nanotubes has the appearance of dryer lint and can be spun and woven into fabrics or wrapped and bound into composites. The material is 100 times stronger than steel, extremely lightweight, and very efficient at absorbing radiation. It is perfectly suited—and may be the only viable material—to cover the hull of spacecraft that could safely transport humans to Mars and back, shielding them from long-term exposure to radiation in space.

Although white graphene is a poor carrier of electrons, it is a superb conductor of heat. As silicon chips become denser, they run hotter, and their speed is limited by that heat. In the near future, white graphene may be built into hybrid silicon chips to draw off that waste heat, allowing them to run much faster. Cohen and Zettl are testing another theoretical prediction: that when tiny balls of carbon known as fullerenes (a.k.a. buckyballs) are placed within the tubes, they can be coaxed into performing as superconductors.

The first white-graphene nanotubes were extraordinarily challenging to forge, requiring temperatures hotter than the surface of the sun. But in 2013, a new method developed by Zettl and others worked like a charm. “They just keep pouring out now,” reports Zettl. “We are doing lots of tests on them, and engaging industry.”

Zettl is one of many loyal colleagues who have worked closely with Cohen over the years. Nearly four decades ago, Cohen took on a young graduate student named Steven Louie. Originally from Guangdong, China, Louie had come to the United States as a child, just like Cohen. When Louie returned to Berkeley after stints at Bell Labs and IBM, another collaboration was born. Today their fifth-floor offices in Birge Hall are only a few doors apart, and when they want to confer about a new idea or a problem, they do not send emails. Using various computational strategies, they both explore the quantum mechanical frontiers of exotic materials such as graphene and buckyballs. “Today,” says Louie, “the generation, transport, and storage of energy is very important. Every five years or so, someone will discover something surprising, and that’s what makes this field so exciting.”

At the dedication of the new discussion room in Birge Hall, Louie called Cohen “a pillar of our physics department,” and a key reason why it was rated number one among 200 universities by the Shanghai Academic Ranking of World Universities in 2014. Also at the dedication was Cohen’s wife, Suzy Locke Cohen, a highly regarded Bay Area art adviser and appraiser who helps place paintings and sculpture in homes and institutions such as hospitals.

The new room wouldn’t have Marvin Cohen’s stamp on it without some art, so he commissioned a painting from the room’s designer, San Francisco architect William R. Glass. The piece, called The Theorist’s Tools, is a bright composition depicting a large window with a view of blue sky and wispy clouds. On a desk in front of the window is an open book, a sheaf of manuscript papers, and a glass cup containing a clutch of yellow pencils spread out like flower stems. Each new pencil is freshly sharpened, each point aiming skyward, waiting for the next idea. 

Sabin Russell is a freelance science writer based in San Francisco. A former reporter for the San Francisco Chronicle, he has also published in California, Health Affairs, MIT Technology Review, and The New York Times. After a Knight Science Journalism fellowship at MIT, he was an in-house writer at Lawrence Berkeley National Laboratory 2010–14.

Share this article