About a quarter century after their discovery, high-temperature superconductors still puzzle the scientific community, but each year scientists deploy new tools in order to understand the phenomenon. One new model uses geometry: Electrons moving in a curved space-time occupied by a charged black hole at the center can mimic strongly correlated systems such as high-temperature superconductors. Theoretical condensed matter physicist Philip Phillips and his colleagues at the University of Illinois, Urbana-Champaign, recently tweaked this model, with interesting results.

Phillips, whose Ph.D. is in theoretical chemistry, says he probably would not have taken such an unusual approach if his background had been more conventional. Science Careers talks to Phillips about how his circuitous route from chemistry to physics prepared him to go down this new research avenue.

The following highlights from the interview were edited for brevity and clarity.

Q: How did you first become interested in science?

P.P.: My parents were in the humanities. I was born in Tobago, and we moved to the U.S. when I was 10. I was very interested in math as a kid. Science was interesting, but I had very bad teachers in high school and so I had no real way of knowing what science would be about.

At the university, I took a chemistry class, and that's when I really became interested in science. So I started taking many more science classes, and I realized too late that my real interest was in physics. I needed one more class for a physics major, and so I had degrees in math and chemistry.

Q: How did you go about choosing your Ph.D. program?

P.P.: I wanted to be a theoretician because math was always my thing, and if I was going to do science, I wanted to apply math to understanding physical problems. [But] I realized, given my limited undergraduate background, that somehow being able to chart a course that was intellectually what would be my focus for the rest of my life was just not possible. So I viewed a Ph.D. as a degree in which I learned how to do research, and the particular problem wasn't something I was deeply interested in at all. My project was on explaining phosphorescence lifetimes in small molecules.

Q: Did you get what you needed out of your Ph.D.?

P.P.: I had an adviser who was an incredibly brilliant person, and he really taught me how to get something done. He gave me this sense of just being able to take on a challenge even if you have no experience with the field.

Q: What was your next step?

P.P.: I got a Miller Fellowship at [the University of California,] Berkeley. The key thing I got interested in was disordered systems, and I started really thinking about many-body systems and phenomena that arise from collective physics, the sorts of things that would define my career.

I just started reading all the papers, and then I defined a new problem that others had not solved that I thought would advance the field. So the problem I was working on at Berkeley was an electron moving in a random array of scatterers. I learned the necessary math tricks to be able to solve this problem, and then I started doing it. That's what I'd learned from my adviser: how to chop something down that is completely new and make progress on it.

As a Miller fellow, I was doing this on my own. It was a big jump from single-particle stuff to, essentially, statistical mechanics, and the mindset was very different. It was painful and a lot of stuff I had to learn, but it was what I knew I wanted to do.

This Science Careers article is a tie-in to Science magazine’s feature on superconductivity.

Q: You then obtained a faculty position at the Massachusetts Institute of Technology (MIT), in the chemistry department. What did you work on there?

P.P.: This problem led me to look at Anderson localization, which is the problem of electrons moving in a random lattice. Solids have a regular array of atoms, and what Anderson showed is that if enough of the atoms are different, the electrons change from being able to move freely to being completely stuck. In one and two dimensions, it's generally been thought that any amount of disorder would lead to localization. But we found the general exceptions. And then I showed that certain classes of conducting polymers could be explained by the examples that lead to exceptions to localization. That application was motivated entirely because I was in a chemistry department.

Q: You then decided to move to Illinois after 9 years at MIT. Why?

P.P.: I was forced out. What I was doing had nothing to do with their definition of chemistry. Some supported me and some didn't. But even if I had gotten tenure, I would have had to move to a physics department.

Q: Were you aware at the time that you were risking tenure?

P.P.: Yes, I was very much aware of this. You have to be honest with yourself. There were no problems in chemistry that interested me. I have always just thought that you should do what you think is important regardless of whether it might work out or regardless of whether or not your colleagues think it will work out.

Q: You obtained tenure at the University of Illinois right away. Do you feel as though you've found your place now?

P.P.: Illinois has been absolutely a gold mine for me, yes. Personality-wise, in terms of the research I'm able to pursue here, in terms of a supportive environment, in terms of my research being central to the condensed matter effort of the department.


Credit: Rick Kubetz, University of Illinois
Philip Phillips collaborated with theoretical particle physicist Robert Leigh, also of the University of Illinois, Urbana-Champaign. Not shown in the picture and also contributors to the work are postdoctoral fellow Mohammad Edalati and graduate student Ka Wai Lo.

Q: So what has been your focus at Illinois?

P.P.: The big problem I have been trying to solve since 1995 is the physics of strongly coupled electron systems. I have been attempting to figure out what it is that the electrons are doing as they interact with one another [within cuprate superconductors]. Our work shows that they form composites, and so once you understand what the composites are, then you can begin to describe the macroscopic properties of the system.

[In 1998, Argentinean physicist Juan] Maldacena made a conjecture in which he argued that there is a relationship between a strongly coupled quantum mechanical system and a gravitational system [that] is entirely classical Einsteinian gravity. So in fact, strongly coupled quantum mechanical systems that are charged are equivalent to a curved space-time with a black hole in it. We showed that if you just introduce some probe fermions and these probe fermions are coupled to the space-time in a particular way, that system looks identical to the normal [nonsuperconducting] state of high-temperature superconductors.

Others have used this mapping before. What we did that was new is that we used a particular interaction between the probe fermions and the black hole that is really irrelevant to the physics of the black hole but changes the physics at the boundary of the space-time [which is where the quantum mechanical theory lives]. No one suspected it.

With such a model, you can just forget about trying to figure out what the basic building blocks are, just go and solve this geometry problem and extrapolate it to what's going on at the surface of this geometry, and you'll see what the quantum mechanical system is doing.

Q: And so have all these years of getting closer to your true scientific interests finally paid off?

P.P.: It takes a lot of daring to invest the time to go and learn this machinery [for geometrizing quantum mechanics] because it's fairly nontrivial stuff, and to think that it has answers for a real-life system is another risk.

I certainly thought that, God, if I were more traditionally trained, maybe I could have known some of the pitfalls of some of the things I've tried in the past. But now it's turning out that it was a good thing. The most important thing about my roundabout way is that I don't have any biases. I'm very open to new approaches and new problems and I don't mind just going and rolling up my sleeves and trying something new. This research is the culmination of what I thought I wanted to do when I was a graduate student.

Philip Phillips C.V. Highlights

1979: Receives bachelor's degree in math and chemistry from Walla Walla College in Washington state

1979–81: Graduate research assistant in theoretical chemistry at the University of Washington, Seattle

1981–84: Miller Postdoctoral Fellowship, University of California, Berkeley

1984–90: Assistant professor of chemistry, Massachusetts Institute of Technology, Cambridge

1990–93: Associate professor, Department of Chemistry, Massachusetts Institute of Technology, Cambridge

1993–99: Associate professor, Department of Physics, University of Illinois, Urbana-Champaign

2000–Present: Professor of physics, Department of Physics, University of Illinois, Urbana-Champaign

Elisabeth Pain is Contributing Editor for Europe.

Elisabeth Pain is contributing editor for Europe.

10.1126/science.caredit.a1100031