News Center
My Hero

Jennifer Warnick

Lead Writer

Written By
Jennifer Warnick

Station Q, headquarters of potentially world-changing quantum computing research, is located just past where the Pacific Ocean meets the sand, up through a grove of palm trees and across a bike path. (Do mind the shirtless college student zipping past on a skateboard wearing only a backpack and swim trunks.)

In some ways, Station Q is not at all what you’d expect from a hub for next-level computing research – there’s a strong California vibe, with world-renowned experts turning up for work in Hawaiian shirts and shorts, and even a nearby room with a shower, a clothes rack full of faded wetsuits and battered, loaner surf boards leaning in the corner for those who do their best thinking while hanging ten.

In other ways, Station Q’s surroundings are exactly what you’d think – equation-packed chalkboards hang in every office, meeting room and hallway; math and science comics taped outside office doors; and an academic air of silence (though there’s an underlying buzz to the place – a feeling of restlessness).

Michael Freedman, Station Q’s director, is stately, fit, and well-tanned. He looks a bit like heroic police chief Martin Brody from the movie “Jaws” (played by actor Roy Scheider) who saves a small coastal town from a man-eating shark. At Station Q, located on the campus of the University of California, Santa Barbara, Freedman and his colleagues from all over the world, both inside and outside of Microsoft, explore the exciting, mysterious, difficult and downright strange space where computer science meets quantum physics.

It’s not hard to picture him striding out of his sunny office, or returning from one of his long, contemplative walks on the beach, to utter Brody’s most famous line from the movie: “We’re going to need a bigger boat.”

Freedman’s central preoccupation for the last decade, quantum computing, could be a bigger computational boat than the world has ever known. It could tackle problems that would take today’s computers eons to solve in the time it takes to grab a cup of coffee. It could have wildest imagination-type applications in fields such as machine learning and medicine, chemistry and cryptography, materials science and engineering. It could allow humans to understand and control the very building blocks of the universe.

It’s no surprise, then, that when they discuss their work on quantum computing, top-notch mathematicians, physicists, computer scientists and researchers – types not typically prone to hyperbole – lean forward in their seats and use terms like “strange and unusual,” “mind-bending,” “exotic,” “magical,” “beyond science fiction” and “world-changing.”

“It could be as dramatic as anyone says,” Freedman said. “It could yield enormous computational consequences. The truth is, we don’t know yet.”

Quantum computing could solve problems that would take today's computers eons in the time it takes to grab a cup of coffee.

He paused. “But one couldn’t have a more exciting playground.”

So is there a relatively easy way to explain quantum computing?

“In short, no,” Freedman said. “There’s a famous quote that says, ‘Everything should be explained as simply as it can be, but not simpler.”

With apologies to both Freedman and Albert Einstein, who is long credited for the above quotation, we shall proceed. Rather simply.

Particles Gone Wild

Thanks to the work of centuries of relentless and brilliant minds, from Newton to Einstein, we have a solid understanding of matter, motion, time and space. But over the last hundred or so years, scientists studying life at atomic and subatomic levels started noticing some inconsistencies with classical physics. Questions and theories started piling up as to how and why particles appear to behave predictably on a large scale (plants and birds and rocks and things) but on a nanoscale it’s … well, particles gone wild. Enter stage right: quantum mechanics, the wild and crazy brother of classical physics.

“If quantum mechanics hasn’t profoundly shocked you, you haven’t understood it yet,” said Niels Bohr, a physicist who in 1922 won the Nobel Prize for his work in quantum theory.

Behaviors that seem impossible to imagine on a human scale are downright pedestrian in nano neighborhoods. At the molecular level, particles in a quantum state can teleport information from one place to another. Particles can also experience “entanglement,” remaining eerily connected no matter how far apart they become (like separated identical twins in some matinee thriller where one bumps her head in Paris and the other, in Los Angeles, absently starts to rub her sore noggin).

As if all of that wasn’t weird enough, in a quantum state, particles can achieve “superposition,” where they exist in multiple states simultaneously. To put it in Fleetwood Mac terms, quantum particles are kind of like Rhiannon, who “is like a cat in the dark, and then she is the darkness.”

What does all of that have to do with computers? In theory, a lot.

The laptop on your desk, the smartphone in your hand, the tablet in your bag (not to mention the theoretical Turing machine and the room-sized, vacuum tube-powered behemoths from which all modern computers descended) work with information in the form of bits. Bits are either a 1 or a 0, and can be arranged in long, artful strings to get computers to perform all manner of tasks, from sequencing DNA to flinging angry little birds at pig-built fortresses.

In quantum circles, these 1 and 0-powered machines are referred to as “classical computers,” to help distinguish between the now and the potential “wow” of computing in a radically new (and possibly exponentially more powerful) way.

“All our machines, no matter how fancy and parallel are basically bells and whistles on top of the original Turing machine,” Freedman said. “There's the classical model, which is the Turing model, and there's the quantum model, and making this transition, jumping from the first to the second, is kind of like getting a peek at the inner workings of the universe, looking behind the screen.”

Classical computing power has grown by leaps and bounds, even as form factors shrink, multiply, and require less energy. Some early supercomputers ran on maybe 13,000 transistors; the Xbox One in your living room contains 5 billion.

As sophisticated as they’ve become, classical computers are limited in their problem-solving prowess. There are some problems so difficult, so incredibly vast, that even if all the computers in the world worked on the problem in tandem they would be sporting that little “I’m thinking hard” hourglass for a long time.

“A very long time – the lifetime of the universe or more,” said Krysta Svore, researcher and manager of Microsoft Research’s Redmond-based Quantum Architectures and Computation (QuArC) group. Svore and her team have developed a software architecture with a program called LIQUi|> (a cheeky mathematical way of writing Liquid) and are testing algorithms on simulated models of up to 30 qubits.

A lot of these virtually unsolvable problems are the tough nuts you’d expect, like computing the ground state energy of a molecule.

“We think a quantum computer could possibly solve these types of problems in a time frame that’s more reasonable than the life of the universe, maybe a couple of years, or a couple of days, or a couple of seconds,” Svore said. “Exponentially faster.”

The difference between quantum and classical computing is all in the approach. Classical computers attack problems like you would navigate a corn maze, those farm-size labyrinths popular in rural areas at harvest time. It proceeds down each long, stalk-lined corridor and at each fork, it picks one direction. If it reaches a dead end, it turns around, finds its way back, and tries another route until eventually it solves the maze (unless, of course, the maze is so massive that examining every route takes the lifetime of the universe).Quantum computers run on quantum bits, or qubits. Because of the bizarre properties of a quantum state, like superposition, a qubit can be a 1 or a 0 – or it can operate as both a 1 and a 0 at the same time. If one qubit, as both a 1 and a 0, can do two calculations at once, then two qubits can do four, and things get exponential pretty quickly.

“It’s like that old story problem from math in school where you offer kids a thousand dollars right now, or to give them one penny today, two pennies tomorrow, and continue to double that every day for 30 days,” said Peter Lee, corporate vice president and head of Microsoft Research.

Most kids want the thousand bucks, Lee said, but the pennies doubled daily over 30 days eventually adds up to more than $10.7 million.

Imagine that same corn maze, but instead of looking for the way out on foot, one turn at a time, you unleash a pack of high-octane, well-trained Tribbles. Those fuzzy, fictional Star Trek creatures would move out in every direction at once and, thanks to their tendency to multiply at frighteningly exponential rates, explore every possible route simultaneously to quickly find the most efficient solution.

The trouble with qubits? They’re fussy. If artists get eye rolls for their back-stage dressing room demands (Versace towels, roses with stems trimmed to exactly six inches, live kittens, Flintstone vitamins), qubits represent another league of diva entirely.

A bump in temperature, a bit of electricity, a stray cosmic wave, a slight jostle – any sort of interference at all (even an inside job – a distraction from fellow qubits) will cause them to “decohere” from their quantum state, at which point the calculation and information are gone. Scaling enough qubits to be useful, doing so in a stable way, and keeping them from falling apart – these are some of the fundamental challenges of the field. And they’re challenges Microsoft and its partners are chasing at a full sprint.

“The problem of coherence is a major focus of our research here,” Lee said. “Every researcher connected to this field dreams of building a quantum computer. We are not trying to build a quantum computer. Our belief is that trying to build a quantum machine by controlling electron spin and using surface codes is like trying to build a computer using vacuum tubes. Labs all over the world can do that, but you’ll never be able to scale up. We’re taking an outrageously hard, unreasonably difficult approach, and if we succeed – and it’s a big if – then we will have a building block for a scalable quantum machine. We have a chance, a tiny chance but a real chance, to completely upend technology and society in a fundamental way just like the transistor did.”

Peter Lee is head of Microsoft Research. Photo by Brian Smale / © Microsoft

It’s easy to be skeptical about whether this quest to harness an all-new kind of computing could yield meaningful results for humankind. But consider this: six decades ago, many people might have been skeptical about the idea that in 60 years, 13-year-old kids would carry touchscreen computers in their pockets or that the average household would have access to a vast, searchable catalog of the world’s information.

“What is so amazing about quantum mechanics is it's the fundamental microscopic language of the universe. It's the way the universe talks to itself, and we don't think that way. We're more like classical computers,” Freedman said. “With this project to develop quantum computing, what we're really doing is making the transition as a species from our devices thinking in this very kind of clumsy, classical model, to our devices thinking in the fundamental language of the universe. So, we're going to be leaving behind this more primitive method of processing information, and we'll move into the quantum realm.”

Now Arriving at Station Q

Freedman is something of a phenomenon. He attended the University of California, Berkeley at age 16, and at 22 had earned his graduate degree from Princeton University. At 36, he won the Fields Medal (the highest honor in mathematics) for cracking the long-standing Poincare conjecture (a problem first posed in 1904 by French mathematician Henri Poincaré).

Craig Mundie, senior advisor to the CEO at Microsoft, in 2012. Photo by Brian Smale / © Microsoft

Freedman specializes in topological mathematics, the study of geometric forms that remain unchanged when bent or stretched. The guy – also a rock climber, swimmer and runner – knows a thing or two about stamina, especially the kind it takes to work on the same problem for years. Several years into working on quantum topology and physics at Microsoft, Freedman was feeling a bit – well, isolated.

“When I joined a company coming from academia, my concern was that I wouldn’t be able to work – that people would be dropping in on me posing various puzzles and problems, and I’d be distracted by lots of small things rather than one large thing,” Freedman said. “I made it clear when I was hired that I expected to have a clear work space. Microsoft respected that to such a degree that after seven years I was kind of lonely.”

In 2004, Freedman approached Craig Mundie with the idea of using topological mathematics to help create a stable platform for manipulating quantum information. Mundie remembers the meeting well.

“He’d been thinking about quantum computing, and had come to understand what it was that people had found so difficult about the prospect of building a scalable, workable machine,” Mundie said.

“If I understand this right, what you’re talking about is adding error correction to a qubit,” Mundie told Freedman.

“Yep, that’s basically it,” Freedman said.

Over the course of several meetings, they discussed the creation of a group to pursue this research. Mundie pushed Freedman to go beyond the theoretical.

“It was actually Craig's idea that the group should interact very strongly with experimentalists – that we should start an external experimental program,” Freedman said. “I was a little bit rattled by the idea that I would be, in some sense, running an experimental program. And I tried to backpedal and said, ‘Well Craig, I'm just a mathematician.’”

“Not anymore,” Mundie said. “Now you’re a mathematician and a program manager.”

Freedman looked to “go where the science is,” setting up shop on the UC-Santa Barbara campus, but wanted to refrain from calling the new group an “institute” or a “center.”

“It’s not a leisurely place, not a place for sipping tea. I thought it would go over much better as a research station. An outpost. Somewhere we’d go to just get the job done,” Freedman said.


Station Q was born, and over the next decade, Freedman became a pied piper of sorts, attracting notables from mathematics, science and computer science – theorists and experimentalists alike – to the cause.

“We were looking for mathematicians and physicists who were extremely flexible in their outlook - who wanted to learn the math or the physics that they didn't know, because quantum computing is really an interface on maybe three different disciplines – mathematics, physics, and computer science,” Mundie said. “So, we wanted people that were either expert in all three, which is the case for some of the people in our group, or at least had an interest and a desire to learn more about the subjects that they weren't experts in.”

These far-flung experts travel to Station Q at UC-Santa Barbara a couple of times a year to catch up with each other, and on the latest research, which is offered in sessions with titles like, “Isolated Superconductor-Semiconductor Nanowires” and “Multi-Junction Devices for Protected Superconducting Qubits.” There is one aspect of the Station Q conference everyone can understand – they have a barbecue on the beach one of the evenings.

Before this year’s conference Peter Lee joked that the Station Q conference sessions are so notoriously complex that after a few minutes into one even he starts to feel a bit thick. “And small,” Lee added. This from the former chair of Carnegie Mellon University’s top-ranked computer science department. Quantum computing is that mind-bending.

Station Q Director Michael Freedman. Photo by Brian Smale / © Microsoft

“Up until recently, maybe two years ago, all of this was so speculative. Even in scientific circles this was considered out in the fringes,” Lee said. “Now I think you could argue that the topological approach has become mainstream. Physicists don’t think we’re crazy people anymore.”

Sankar Das Sarma, a distinguished professor and theoretical condensed matter physicist at the University of Maryland, certainly doesn’t. He’s been collaborating with Station Q for a decade. Buttoned-up with perfect posture, Das Sarma is an initially serious-seeming guy who it turns out is to his field what Neil DeGrasse Tyson is to the cosmos – the man can explain even the most complicated ideas about quantum physics like he’s recapping last night’s “Game of Thrones” episode. Quantum computing isn’t a small part of the way the universe works, he said.

“Everything is quantum mechanics,” he said. “We’re just trying to make the move from using quantum physics passively to actively using quantum laws to do something really different and revolutionary.”

His analogy? Airplanes.

“There’s this plane flying through the sky. If you drop a coin from the plane, it still falls – it’s not like the laws are different for the plane – it’s just that in flying the plane you are controlling those laws of classical physics and thermodynamics to your advantage with things like jet fuel and controls. We understand Newton’s laws so well that we can use them to our advantage. This is what we’re trying to do with quantum computing.”

Hunting Quasiparticles

In 2000, physicist Alexei Kitaev (then at Microsoft Research) proposed that a mysterious quasiparticle known as a Majorana could be used in quantum information processing, showing that Majoranas located at opposite ends of a quantum wire could effectively create a topologically protected qubit. Five years later, Das Sarma – along with Freedman and Chetan Nayak, Station Q’s other main leader – co-authored a paper suggesting an experimental proposal for creating a topologically protected qubit using something called the “fractional quantum Hall” system (two-dimensional electron gas in a strong magnetic field) along with a similar quasiparticle. These important discoveries pointed to a promising new direction for protecting qubits, and therefore getting them to behave. After all, qubits working together in harmony is fundamental to getting them to compute.

As they work together to compute information, qubits are a bit like a perilously balanced pyramid of cheerleaders. If even one qubit is disturbed while the particles are passing information back and forth during a computation, the whole thing comes tumbling down. And when it tumbles, or decoheres, all of the information that was whizzing around is lost for good. But a qubit that is topologically protected by one of these mysterious quasiparticles could make it so a disturbance in one has no effect on the others – or the information being processed. Imagine being able to remove the cheerleader right in the middle with no effect on the overall integrity of the peppy pyramid. Similarly, this type of quasiparticles could in theory help create a much more stable quantum environment. The only problem? At that time, and for a century beforehand, the existence of Majorana particles was purely speculative.

This work attracted the attention of a number of physicists, including Charlie Marcus, a condensed matter physics professor at the Niels Bohr Institute in condensed matter physics. At a Harvard luncheon where they met by chance, Freedman explained to Marcus (an experimentalist who was then teaching at Harvard) why it was basically impossible to set up an experiment. Marcus was undeterred. “It doesn’t sound that hard,” he told Freedman that night. “Maybe we should think about doing it.”

“It was more than just wanting to work on that physics problem, the fractional quantum Hall effect,” Marcus said. “Michael is such an astoundingly warm, clever, attractive personality that I just sensed immediately that he was a guy I also wanted to be friends with. So we developed ideas about how to make an experimental program to allow some of the ideas to be tested.”

[Station Q] is not a leisurely place, not a place for sipping tea.

A few years later, when new proposals for realizing Majoranas based on nanowires were making the theoretical rounds, Marcus said, ‘Let’s get Leo involved. He’s actually good at the stuff.’”

“Stuff” being nanowires and superconductivity, and “Leo” being Leo Kouwenhoven, a professor of physics at Delft University of Technology in the Netherlands, and Marcus’ longtime friend-slash-rival (“rival” used here mostly in a “competitors with mutual respect vying for Iron Chef” kind of way). They joke that they’re the Laurel and Hardy of physics.

“We share recipes, but each of us still wants to be the best cook,” Marcus said. “It’s fun to compete. We’re trying to advance the field.”

Researchers at the University of Maryland led by Das Sarma and a group of international physicists working in parallel provided the experimentalists with a so-called recipe for how they might reveal the existence of Majorana particles, and Marcus and Kouwenhoven set out to follow it.

No one knew what to expect. Kouwenhoven and his small team had a few failures, made a few adjustments, and in late 2011 believed they’d detected Majorana quasiparticles. The professor kept the findings under wraps for a while, trying to find any other explanation for what appeared to be a scientific smoking gun. Finally, in April 2012, Kouwenhoven made the announcement. His team had successfully detected compelling evidence of a Majorana particle. It made the cover of Science.

“This is the first step toward a topological qubit,” Kouwenhoven said. “First we had to identify a physical system that has Majorana properties. The next phase is to show that Majoranas have topological properties. The person who demonstrates that – that’s huge. The last thing, demonstrating the DNA of particles that makes them relevant for quantum computing, will also be huge.”

Physicist Leo Kouwenhoven (left) and Charlie Marcus at the 2014 Station Q conference. Photo by Brian Smale / © Microsoft

It may seem like one small step for qubit-kind, but it was a landmark finding that kicked off a topological quantum computing craze in the physics community, said Microsoft’s Peter Lee.

“It’s not definitive proof, but very strong evidence, and several other experimental physics groups around the world have since come up with similar results in their own independent experiments,” Lee said.

Station Q researchers will continue to try to do very difficult things, he said, including continuing to pinpoint the existence and characteristics of Majorana particles; trying to detect particles called anyons and explore how those particles might make calculations; finding a way to “braid” strings of anyons through time and space to create stable qubits and therefore quantum properties; and exploring ways to apply topological effects to make qubits more robust.

Freedman and his fellow researchers use donuts to help explain why Station Q is pursuing topological effects in the field quantum computing. If you have a donut with a hole in the middle, it’s still a donut with a hole in the middle even if you take a large bite out of it, he said. Now imagine a quantum donut made of qubits, undisturbed and blissfully performing exponential computations. If the qubits had topological properties, it wouldn’t matter if there was a bite taken out of the quantum donut – the functionality of the donut could be preserved rather than a disturbance causing the whole thing to decohere, or fall apart.

“Control of quantum information would be a fantastic milestone, but our primary motivating factor is science. We’re a big research organization, one of the things we’d like to do is make our mark intellectually in one of the most interesting and promising areas in science,” Freedman said. “We’re in early days. But the science is beautiful and first-rate, as is the mathematics, and it certainly helps to have in the back of our minds we could be developing the foundations for a new kind of technology – sort of a post-silicon age.”

From Here to Infinity

The quest for quantum computing is a bit like the tortoise and the hare, Mundie said. Groups around the world are trying to find ways to do quantum computations, and some have even built machines with quantum properties, but those may suffer the same fate as the vacuum tube computers of old.

“They may be like the hare in their ability to get a handful of qubits to do something, and to demonstrate that quantum computation is possible, but they may then hit a wall in terms of their ability to scale up,” Mundie said. “Station Q is a treasure trove of research. I think we’ve shown leadership in taking a more soup-to-nuts approach to quantum problems than any other organization I know, which is not in any way meant to diminish achievements others have had in contributing fields.”

We could be developing the foundations for a new kind of technology – sort of a post-silicon age.

Station Q’s strategy, to build a quantum computer based on “topological degrees of freedom,” is theoretically harder to get off the ground initially, but if and when they get it working, it will be very scalable.

“Our pursuit is not strictly academic in nature. We dream a dream this will one day inform our product strategy, and be of utilitarian and economic importance,” Mundie said. “Because of that we care about getting on a path that would give us the quickest ability to do something that is economically important as opposed to just academically important.”

The economic implications could be staggering. The ability to harness quantum properties could usher in a second-coming of the computing age, one with vastly more power and speed than the silicon era.

A genuine quantum offering would, in theory, utterly dwarf the power of today’s computers, making all sorts of problem-solving and applications within the realm of wild speculation – factoring very large numbers, new frontiers in cryptography, synthesizing better drugs, creating designer materials and energy sources, discovering new particles and elements, bolstering artificial intelligence, better understanding the universe – the list goes on.

Still, researchers say quantum exploration is a bit of a boomerang. They’re throwing it with as much force as they can, but there’s no guarantee of when or how it will come back (or whether it will at all).

“If you ask us a lot of questions about what we’re working on right now, and what is it all good for, in some cases we’ll say we want to make a qubit, or make a quantum computer, but in many cases we have absolutely no idea,” Marcus said. “What we discover may sit there for a while, and later become relevant. The truth is always relevant eventually. Dormancies frequently follow important discoveries, which will be brought back up at their right time. This story is one of the value of basic research.”

Das Sarma agrees.

“What can all this do? My imagination is not powerful enough to predict,” he said. “I’m not being humble. We have no idea what a quantum computer could do. Factor numbers? Make better drugs and materials? These are all possible applications. There’s a good case to be made that quantum computing could change the world.”

Physicist Sankar Das Sarma has been working with station Q for over a decade. Photo by Brian Smale / © Microsoft

Bottom line: nobody knows. Das Sarma said being able to see applications of a quantum computer now would be like Isaac Newton waking from the dead and seeing that airplane harness (and defy) his laws.

“He’d immediately die again,” Das Sarma laughed. “I’m serious! At one time the head of IBM said he couldn’t imagine a scenario where ordinary people would ever need to have access to a computer, and now in my family, between us, we have 10. A quantum computer will do things we cannot possibly know now, because that’s what happens.”

Mundie said right after he shook hands with Freedman to seal the deal on creating Station Q, he asked him, “Hey, if you ever did build one of these things, what’s the first thing you’d do with it?

Without hesitating, Freedman replied, “I’d use it to design the materials from which to build its successor.”

Like using the last wish granted by a Genie to ask for more wishes?

Mundie laughed. “Yes, I thought it was a fundamentally deep answer.”

Though there is much excitement about the destination, the potential of mastering the tiny cosmic building blocks of the universe, there is also beauty in the journey, Marcus said.

“We live with the belief that what we see is all there is. The world is a much richer place than your first observation tells you,” Marcus said. “As we peel back more layers, perhaps many things will still be hidden from us, but we’ll have a deeper understanding of the world. To think that we are smart enough to engage in this process – to slowly unravel things which are true but have been veiled – should make us very proud to be human.”