Professor Philip Phillips was trained as a theoretical chemist but is among the leading thinkers today in the field of theoretical condensed matter (or solid state) physics. Indeed, it is this very background in chemistry that he believes enabled some of the new thinking he has brought to his field. His research seeks to explain experimental observations that challenge standard theoretical paradigms in electron transport and magnetism in solid state physics. He has also made important contributions to the understanding of conductivity and superconductivity, which are driving technological developments in areas like electronics.

NIHERST interviews Philip Phillips

Q: Your childhood was split between Trinidad and Tobago and the United States. How did those early years shape your sense of identity and choice of career?
A: I was born in Tobago. I have four siblings. Both my parents had studied history and were teachers. We moved to Trinidad and I attended San Juan Primary School before we left for Boston when I was 10. My parents were very authoritarian, beyond the strictness of the typical Caribbean upbringing of that era. I certainly wasn’t free to be myself. But what I did know from a very young age was that I could privately think whatever I wanted to! So I cultivated my imagination, and maybe that’s when the seeds of becoming a theoretical physicist were sown.

Q: What was school like? Were you a good student?
A: In Trinidad, I definitely learned discipline. I was very good at math. English, too, but I really wanted to perfect math. My older sisters helped me advance beyond my grade by teaching me square roots or whatever they were learning. So I found school very engaging because, once again, it was all about the mind. When we moved to Boston, then later to Walla Walla where my father went to teach, I did just okay at high school. Not as well as in Trinidad because schooling there was so by rote—a lot of memorization and repetition. I was a curious student but I didn’t really catch on to the American system. Math and English I would always do well in but the rest I didn’t care too much for.

Q: Not even the sciences – physics, chemistry and biology?
A: Of those, I was best in biology in high school. Believe it or not, I was absolutely horrid, and I mean horrid, in chemistry and physics! I got Cs and Ds if I was lucky. Nothing in chemistry made any sense and physics was even more of a mystery. It wasn’t a problem of being unable to do the math. I just couldn’t do the physics.

Q: Yet you went on to Walla Walla University and majored in chemistry and math and did many physics courses. How did you make that giant leap given your poor understanding of those subjects in high school?
A: I didn’t really know what I wanted to do. I wasn’t sure I wanted a career in math. Then the summer after my freshman year at college, I read a workbook on chemical bonding and I found it quite easy. I then took chemistry from a professor who was supposedly the toughest teacher and I did very well. We went over the periodic table of elements. I had never realised there was any kind of structure or theory behind it. The elements are structured according to their similarities, and the similarities are predictable, based on the number of orbitals, protons and electrons they have. The numbers mean something and they predict their function. That was revelatory and revolutionary to me, seeing there was structure, order and predictability in this vast array of stuff on the periodic table! That’s when I knew I wanted to be a scientist. Not an experimental scientist but a theoretical scientist. Every question I asked, I didn’t think of an experiment that I needed to do; I thought about the mathematical equation I would use to gain an understanding. The following year I took physics – the other thing I wasn’t good at in school – and that now didn’t seem very hard. And I realized why. In college, we were given deep explanations as to why things are, you weren’t just left alone with this “cookbook” type of approach of high school. I need to know the conceptual basis before I can really access something.

Q: You did your PhD in theoretical chemistry. Can you explain briefly what that is?
A: Theoretical chemistry is a branch of chemistry where physics enters more prominently. I was in a group that attempted to predict the properties of single atoms and molecules using computer programs. But I didn’t care about the properties of a single molecule of oxygen or if oxygen is similar to sulphur etcetera from a physics perspective. You don’t see a single molecule in nature. You see a liquid, air, and things are very interactive.
Theoretical chemists attempt to use the abstraction from a single molecule as a way of understanding the whole. That approach assumes no new physics happens when the molecules come together but, of course, a lot does happen. That’s how you get a solid or a liquid. You can’t go from any understanding about a single molecule to predicting a liquid. So ultimately, I didn’t find this interesting.

Q: But you still did your PhD in it and not in physics?
A: Yes, I used the PhD as a degree in which I learned how to do research. I figured out that was the real goal of a PhD. For some, a PhD is career defining. They end up working in that area the rest of their lives. I wasn’t going to do that.

Q: You took up a fellowship at University of California, Berkeley. What did you focus on there?
A: I started doing statistical mechanics which was a very difficult jump to make from chemistry. It’s a branch of physics which tries to predict what the properties of collections of atoms are from the basic interactions between them. Statistical mechanists study polymers (substances whose molecules have high molar masses and are composed of a large number of repeating units), rubber, the elasticity of solids, things like that. The problems there have a more conceptual basis. I’m interested in things that have deep phenomenal issues, so I found I could explore that through solid state physics.

Q: What is solid state physics?
A: Its goal is to understand the electronic and magnetic properties of solids. At Berkeley, I started taking the quantum background I had from my PhD and applying it to collections of things. Chemistry departments were realising they couldn’t ignore solid state physics as a lot of things were happening in it and they wanted chemists who could straddle both fields. There were actually quite a few of us. So I got a job at Massachusetts Institute of
Technology (MIT). But the amount of straddling that was tolerated tapered down so I converted from being a chemist to a solid state physicist. For me, that was where the deep ideas and problems were.

Q: What would be an example of a deep idea for you?
A: Well, super-conductivity for one – spontaneous symmetry breaking and localisation. Localisation was my first physics research problem. The fact that you could trap electrons by introducing randomness and changing their environment is a big idea. One area I started working on was conducting polymers. I came up with a general exception to Anderson’s Localization Theory. Phil Anderson had won the Nobel Prize partly for this theory, which predicts that conduction in one dimension is impossible because of random defects. An orderly array has the same kind of atoms at every site. If you introduce any disorder, the electrons won’t be able to conduct i.e. move from one end of the wire to the other. Prior to that, people thought that because electrons are waves, they’ll always conduct electricity. So it was a surprise that just disordering the system would trap electrons.

But in the early 1980s, a class of polymers was discovered that conducted electricity. These are very thin, one-dimensional strands of polymers. I thought, if these polymers conducted electricity so well, then how on earth are they doing it? The accepted theorem says they shouldn’t be able to. Defects in the polymers would disrupt the flow. If it were true, there had to be exceptions to the theorem, and I came up with one. That’s the work I did at MIT. It’s called the random dimer model, explaining the conduction of some polymers. People initially thought that my theory had to be wrong. It went from being outrageous to almost trivial, then accepted virtually overnight. You could work through the math and it was so simple. People were kicking themselves because they didn’t think of it. It’s still my most cited work. The original paper has over six hundred citations.

Q: After nine years, you left MIT to take up a teaching and research post in physics at the University of Illinois. How
was that transition for you?
A: I never in my wildest dreams thought that I could end up there because it’s the top solid state physics department in the world. This is the department that Nobel laureate John Bardeen built. He invented the transistor (a semiconductor device used to amplify or switch electronic signals and electrical power) and developed the theory for superconductivity. So you can’t just come to Illinois unless you’ve really done something. It was the perfect place for me. We try to explain experiments in a completely independent way, not wedded to any particular school of thought.

Q: You’ve done some seminal work at Illinois on Motts’ theory of insulation as well as on superconductivity. Can you give an overview of your contribution in those areas?
A: The big thing I’ve done there is Mottness. The Mott problem became central to solid state physics in 1986 when high-temperature superconductivity in copper oxide ceramics was discovered. Superconductivity is electrical transport without any resistance. At ultralow temperatures, most metals superconduct. The key advance in 1986 was that the transition temperatures for superconductivity rose above the temperature where air – nitrogen – liquefies. Liquid nitrogen is cheaper than beer so the refrigeration problem facing traditional superconductors was solved. Theoretically, however, the new materials were a complete puzzle because they were a special type of insulator called a Mott insulator. Traditional insulators insulate because all the electronic states are completely full. Mott insulators insulate even though there are lots of empty electronic states. Mott theorized that this happens because the electrons are so repelled from one another they just stay put. Slater, on the other hand, argued that Mott insulators insulate because a transition occurs that results in the electronic states being filled. I invented the term Mottness to help
settle this debate. If Slater is right, then all properties of Mott insulators could be explained by the transition he had in mind. Mottness refers to whatever is not accounted for by the transition that Slater proposed. In essence, Mottness refers to the explanatory residue which is left over if one attributes Mott insulation to the Slater mechanism. Experimentally, there is a lot left over and it is central to explaining what is going on in high-temperature superconductors before they superconduct.

To explain Mottness, I have had to use many concepts and methods from particle physics. It turns out that the problem of Mottness is the problem of strong interactions. In strongly coupled systems, the whole exhibits properties not seen in its constituents. An example of strong interactions is the hardness of rubber as a result of vulcanization which is the chemical process for converting natural rubber or related polymers into more durable materials. After this process, individual strands of the polymer that make up rubber become intertwined in a mesh-like structure. That mesh-like structure is not a property of any individual strand but an organisational principle of the whole. Protons are made up of tiny particles called quarks. They never occur free in nature, just bound inside a proton. Why? The strong interactions between electrons in a Mott insulator is a similarly deep problem. To explain this problem, I have had to invoke excitations called unparticles and even use methods from string theory. The similarity all these problems share is that some organisational principle emerges from the strong interactions, which results in some high-order structure. In the case of rubber, it’s the cross-linking that creates a material capable of driving a car on. I think the strong interactions in a Mott insulator create new entities which have no particular mass. All particles have a well-defined mass. Unparticles have no definite mass and hence no particular energy. Precisely how such entities superconduct is what I am currently working on.

Q: Does it matter if a complete theory of superconductivity is not arrived at? If we already have, and will continue to have, materials and technologies for superconduction, is the theoretical work so important?
A: Yes, because if you understand the theory then you can start understanding new things. Understanding quantum mechanics led to the revolution in nanotechnology. We’re discovering new superconductors but we still don’t know how they work. What we need is an agreed-upon theory. Right now, we have lots of statements that can explain this or that fact but no general principle that unifies all the experimental observations. If we unlock the guiding principle, that will help lead the experimental work. Answering fundamental questions about nature has always led to technologies that could not have been predicted. It is the deep understanding of nature that has propelled human civilisation forward.