Is friction real? Once, with the quiet certainty of someone who just stayed up all night in the company of equations describing concrete, my college roommate told me that friction was made up.
Now, I’m pondering her words as I stare at six yterrbium atoms. They are blue and dancing, projected on the wall of a small room off a long hallway at MIT. Lasers and electronics march all over a wide tabletop, climbing up into the ceiling and slithering down to the floor. I’m about to learn that everything I thought I knew about friction is a 14th-century work of fiction—and that the truth is stranger by far.
I’m in the lab of Vladan Vuletic, a professor of physics here, where two of his graduate students are feeding electrical current through a circlet of aggressively coiled wires into a shoebox-sized, airless vault, instructing the ytterbium atoms to move in unison—in time with swing music, even.
Vuletic and his lab group spent years setting up this maze of a room in order to study technology so new we’re not quite sure if it really exists yet: quantum computing. But when he realized that their experiment could do much more, his curiosity sent him on an unexpected detour. “We could study friction in a way that was not possible before, namely, have direct access to looking at each atom individually.”
Friction is a simple word that glosses over a complex phenomenon arising from a dizzying array of interactions. “Friction is a very elusive thing. It’s not something you can touch, but you always feel the effect,” says Ali Erdemir, a senior scientist at Argonne National Laboratory who has spent decades figuring out how to reduce friction losses in transportation. Friction gives and friction takes away. It ensures that our shoes don’t slip and our vehicles stop on command. But friction also eats up roughly one third of all the fuel we burn in our cars, and deep underground, friction between bits of the earth’s crust decides when and where an earthquake will occur.
Tribologists, the clan of scientists and engineers who study interacting surfaces, think on scales ranging from atoms to airplane wings, and their efforts have huge potential payoffs. In transportation alone, researchers think that reducing the energy lost by surfaces rubbing against each other in engines could save 1% of the all the energy used in the U.S. alone, says Robert Carpick, a professor of mechanical engineering at the University of Pennsylvania.
For tribologists, the experiments going on right now in Vuletic’s lab could offer a fresh window into a force that’s almost as poorly understood as it is ubiquitous. “In some ways there’s more fundamental physics in our understanding of black holes light years away from us than there is about the friction between our feet and the ground,” Caprick says.
Laws and Loopholes
Most people’s first encounters with the scientific side of friction are brief and quickly forgotten. Carpick, for example, had no idea friction was a subject of active research until he started working in a tribology lab as a graduate student. Jaqueline Krim, who heads a nanotribology lab at North Carolina State University, says that just two basic laws about friction, wedged into an introductory physics course, comprised “almost 100% of what I learned up to my Ph.D.”
In fact, those basic two laws go back a long way. “Leonardo da Vinci and the other guy—whose name I don’t actually remember—wrote down the laws by 1600,” says George Smith, a historian of science at Tufts University and mechanical engineer. Da Vinci worked out his rules 200 years before the other guy—Guillerme Amontons—but da Vinci never published. Amontons printed up his laws in 1599, died shortly thereafter at the age of 42, and all but disappeared from history.
Here’s what Amontons’s laws say: Imagine dragging a reluctant elephant across a parking lot. Suppose this hypothetical pachyderm is stubborn enough to keep all four legs locked in place and all four feet touching the pavement. Once you overcome inertia to get the beast moving, all your effort goes into fighting the friction between the elephant’s hooves and the asphalt. Amontons’s first law says that the friction force is proportional to the force of the pavement pushing against the weight of the elephant. (In physics, this is called the “normal force” because it’s normal—that is, perpendicular—to the surfaces in question.) So if you stack a second elephant on top of the first, you get twice as much friction because you have twice the normal force. (Though the normalcy of the stacked elephant situation is admittedly debatable.) The second law states that friction doesn’t depend on how much area is in contact. So if your elephant daintily lifts up one of its front legs, and one of its back legs, the friction doesn’t change, even though there’s only half as much hoof area touching the ground.
Amontons’s laws do a reasonably good job of describing many everyday situations, but they are nonetheless fiction. They fall short because they don’t really tell us anything about what’s going on between two sliding surfaces. The closer we look, the more loopholes tribologists are finding in Amontons’s laws.
For example, let’s take a second look at the second law. If friction comes from interactions between two surfaces, then wouldn’t more surface mean more opportunities for things to catch and snag against each other and thus more friction? “This is something that always intrigued me, you know,” Vuletic says. “It turns out that even this is not perfectly well understood.”
Or take this other example: Which is easier, dragging a box across an ice rink or a soccer field? You might expect that smoother surfaces like ice always slide more easily than rough ones like grass. But this is not always true. If you take two copper surfaces and polish them to perfection, then the copper refuses to slide at all. “When the atoms in contact are all of the same kind,” explained the physicist Richard Feynman in one of his lectures, “there is no way for the atoms to ‘know’ that they are in different pieces of copper.”
Yet another unsolved tribology mystery involves a Soviet physicist named J.W. Obreimoff, who in 1929 was using a Gillette razor to slice rock the hard way. He cut into a thin sheet of mica, blade parallel to the glittery surface. As he sliced Obreimoff saw what he described as a “splash of light.” To this day, neither Amontons’s laws nor any other description of interacting surfaces can explain the phenomenon, says Seth Putterman, a professor of physics at UCLA. Yet it’s everywhere. The same physics is at work when you crunch down on a wintergreen-flavored Lifesaver candy and see sparks or when a cat’s fur crackles with static electricity after it walks across carpet. “For sure we don’t understand the cat’s fur,” Putterman says.
A Rough Place
Our partial ignorance may really be an issue of scale. If you zoom in enough, the seemingly smooth surface of an ice sheet or mirror would resemble a mountain range. “Atomically speaking, there’s no such thing as a flat surface,” says Michael Strano, a professor of chemical engineering at MIT. When you slide one surface over another, it’s like you’ve turned the Himalayas upside down and started dragging them across the Rocky Mountains. The peaks, called “asperities” in tribology lingo, bump into each other. Each time they stretch, compress, or break off saps energy from the motion.
The rough nature of smooth-looking surfaces could help explain why the second law of friction suffices for macroscopic objects but breaks down if we zoom close enough. Most of what we measure as an object’s surface area (say, the elephant foot) doesn’t interact with the other surface (say, the pavement). In fact, only a few atoms at the tops of the asperities in the foot get close to the tops of the asperities in the pavement. These are the only atoms that “actually see each other,” Erdemir says. “They are intimately interacting.”
If we could master those interactions, we might be able to get rid of friction.
Researchers theorized in the late 1980s about how to eliminate one type of friction, known as stick-slip. Stick-slip friction happens when the peaks of one surface nestle down into the valleys of the other and get stuck—until you apply enough force to coax them up and out. In many cases, it’s the dominant frictional effect at atomic scales.
The trick to overcoming stick-slip is to induce apathy, convincing the two surfaces not to give a damn if you move them across one another. Such surfaces are called “incommensurate.” To picture incommensurate surfaces, suppose we papier-mâchéd over one-inch round marbles spaced exactly one inch apart. To make the second surface incommensurate with the first, we papier-mâché over more marbles to make a surface that can’t mesh with the first one . That means making the space between the marbles in the second surface different. (Not just any spacing will do—if the new bumps are exactly two inches apart, then the surfaces will still fit together, with every other peak corresponding to a valley. For the surfaces to be incommensurate, the ratio of the spacing must be an irrational number, which cannot be written as a ratio of integers. A ratio of π would work, but ratios of 17 or ⅓ would not, because then every 17th atom or every third atom would line up with atoms in the other surface.)
If you build two incommensurate surfaces, no matter how you shove them around, “you’ll always have some fitting and some not fitting,” says James Hone, a professor of mechanical engineering at Columbia University. If the surfaces can interlock, they’ll prefer the stuck-together arrangement. But for incommensurate surfaces, apathy sets in. “Then the system doesn’t care if it’s moving sideways,” he says, so you don’t lose energy as you move. If done right, the two incommensurate surfaces might slide past one another with vanishingly low friction. Such surfaces are especially intriguing to the materials scientists, physicists, and engineers that have spent the last 25 years trying to observe frictionless sliding, a phenomenon known as superlubricity. Some argue they’ve already found it.
In an airless chamber in the center of the lab, Alexei Bylinskii, a graduate student in Vuletic’s group, uses electric fields to corral a handful of ytterbium ions into a space the size of a matchbox. By changing the electric current flowing through the maze of wires, they can carefully pull these atoms over a surface below and measure how much friction the atoms feel.
This lower surface, which is designed to be incommensurate with the string of ytterbium ions above, is called an optical lattice. It is made out of light but that doesn’t mean the surface is an illusion. By bouncing light between two mirrors, the group creates a standing wave of light—imagine the peaks and troughs of a frozen ocean wave. These peaks and troughs correspond to points of higher and lower energy for the ytterbium ions, which want to move down into troughs and away from peaks. From the ion’s point of view, this landscape resembles the high and low points on the surface of a material like copper—but the scientists can control the shape and size of the optical lattice far more precisely than they can control the surface of a physical chunk of metal. When Vuletic’s lab rigged the spacing of the ions to be incommensurate with the spacing of the optical lattice below, they observed a dramatic reduction in friction.
Even outside this pristine vacuum chamber, researchers have created systems with incredibly low friction. Ali Erdemir and colleagues at Argonne National Lab recently created a surface coating that resembles miniscule ball-bearings. The “ball-bearings” are actually tiny diamonds, wrapped up in a wispy layer of graphene to produce two incommensurate—and incredibly slippery—surfaces. Erdemir calls the work a clear example of superlubricity.
But some tribologists argue the term is misapplied—“you might call it very good lubricity, not superlubricity,” Carpick says. In physics, the prefix “super-” typically applies only in extreme situations. Sokoloff explains that when researchers observe superconductivity, current flows unhindered because “electrical resistance really does go down to zero.” Similarly, when liquid helium exhibits superfluidity, its viscosity vanishes, allowing the stuff to eerily climb up and over the walls of its container. So far, superlubricity experiments have demonstrated very low—but not actually zero—friction.
The terminology dispute hints at something deeper than a quibble over nomenclature. The theoretical picture of superconductivity relies on quantum mechanics. Superfluidity also defies classical physics. So will quantum mechanics help us understand where to look for “true” superlubricity?
Not necessarily, Sokoloff argues. “Right now, according to the way we understand things…you’re probably not going to see truly zero friction,” Sokoloff says.
Scientists don’t yet know what role quantum mechanics might play in friction on the atomic scale. Vuletic’s lab is working on cooling their experiment down to just a hair above absolute zero, where they hope to see ytterbium ions quantum tunneling—moving through the peaks, rather than over them. They want to see how this quantum tunneling affects friction, an observation that may help us understand friction at larger scales. But it’s not a done deal, Bylinskii says. “Whether friction in the real world depends on quantum mechanical effects, that’s an open question.”
If answering that and other questions would be helpful to traditional mechanical engineers, it would be a breakthrough for nano engineers. At large scales, we’ve come up with shortcuts to make friction less destructive. Take a car tire. On average, every revolution on pavement wears off one layer of atoms, Vladan Vulutec says. “For a tire it doesn’t matter, because there’s billions and billions of layers until you have a millimeter or centimeter of loss of profile.”
But as researchers design devices that are only 100 or even ten atoms thick, and losing even one layer of atoms is a pretty big deal. “Nanoscale stuff is all surface,” says Hone, the Columbia nano engineer, “and so once you contaminate the surface, you’ve changed what it is.”
Back at MIT, Strano’s group is interested in scaling up nanoscale discoveries about friction and other phenomena to make exotic materials for safer, lighter cars and airplanes The potential applications are huge: Ali Erdemir of Argonne estimates that mitigating friction losses in transportation alone could save an estimated $500 billion in fuel costs and 800 million tons of CO2 annually.
Mastering friction could also help make cars safer. When a car crashes, several thousand of pounds of mass that were moving suddenly aren’t anymore. As the energy of motion dissipates, the frame of the car—not to mention the occupants—often crumple up. “You’d like to flow that energy in a certain way, and you’d like it not to go to you,” Strano says. Controlling atomic-level friction could help design a material that’s rigid in most cases but bends easily when pushed from a certain direction, allowing designers to carefully orchestrate how the frame of a car deforms in the event of an accident.
All of the researchers I talked to say they’re a long way from completely eliminating or even expertly controlling friction in the messy world outside the laboratory. Strano points out that, in general, researchers observe amazing properties on atomic scales, but they have made slower progress in advancing tantalizing technologies like ultra-efficient engines or futuristic airplane wings.
But that hasn’t stopped them. “It used to be that friction was okay if your car wasn’t wearing away,” says Krim, the nanotribologist. “Our world is less tolerant now of waste.”