Paul Chaikin

Notice: We are in the process of migrating Oral History Interview metadata to this new version of our website.

During this migration, the following fields associated with interviews may be incomplete: Institutions, Additional Persons, and Subjects. Our Browse Subjects feature is also affected by this migration.

We encourage researchers to utilize the full-text search on this page to navigate our oral histories or to use our catalog to locate oral history interviews by keyword.

Please contact [email protected] with any feedback.

ORAL HISTORIES

Credit: NYU Shanghai

Interviewed by

David Zierler

Interview date

July 22, 2020

Location

Video conference

See catalog record for this interview.

Usage Information and Disclaimer

Disclaimer text

This transcript may not be quoted, reproduced or redistributed in whole or in part by any means except with the written permission of the American Institute of Physics.

This transcript is based on a tape-recorded interview deposited at the Center for History of Physics of the American Institute of Physics. The AIP's interviews have generally been transcribed from tape, edited by the interviewer for clarity, and then further edited by the interviewee. If this interview is important to you, you should consult earlier versions of the transcript or listen to the original tape. For many interviews, the AIP retains substantial files with further information about the interviewee and the interview itself. Please contact us for information about accessing these materials.

Please bear in mind that: 1) This material is a transcript of the spoken word rather than a literary product; 2) An interview must be read with the awareness that different people's memories about an event will often differ, and that memories can change with time for many reasons including subsequent experiences, interactions with others, and one's feelings about an event. Disclaimer: This transcript was scanned from a typescript, introducing occasional spelling errors. The original typescript is available.

Preferred citation

In footnotes or endnotes please cite AIP interviews like this:

Interview of Paul Chaikin by David Zierler on July 22, 2020,
Niels Bohr Library & Archives, American Institute of Physics,
College Park, MD USA,
www.aip.org/history-programs/niels-bohr-library/oral-histories/46756

For multiple citations, "AIP" is the preferred abbreviation for the location.

No idea. No idea. I knew Caltech was really good, you know, it was really well known and stuff like that and was really good in science and stuff, but for example, I—no, I didn’t know Feynman and Gell-Mann—the first week I was there—so I took a bus across the country. Went to Caltech and I was there and things were a little bit strange and then I went to—they had a faculty party.

So the house I was in had faculty over for some party right? And what I really distinctly remember is I hadn’t noticed before this that the accents were different, that the California accent was not the same as the Brooklyn accent, right? Suddenly across the room, I hear somebody speaking absolutely perfect unadulterated English. And that, of course, was Feynman. And that’s the first time I met Feynman.

Zierler:

That’s hilarious. He must have learned how to stop speaking Brooklynese at some point himself.

Chaikin:

No. He cultivated the fact that he didn’t lose his Brooklyn accent. This is before he had a Nobel Prize, so he was well known to people that knew him well, but not to the general public. So that was an experience.

Zierler:

What were your impressions of Caltech when you first arrived and you realized it wasn’t on the beach and it wasn’t surfing all day long?

Chaikin:

Well, I think the impression I got wasn’t right away, but the real impression I got over years—which started as soon as I got there—is that there were people that were orders of magnitude smarter than me. Not, you know, just a little smarter than me, OK? I never thought I was tremendously smart. Like, I didn’t graduate Valedictorian or anything at Stuyvesant. I did OK, but I wasn’t the top guy.

But I got to Caltech and there were clearly people, students, that were smarter than me and there were clearly, you know, faculty and not just smarter, but students ten times smarter. And there were faculty than were ten times smarter than that and then there was some faculty that were ten times smarter than that, right? So I really got humbled by it, right? At some point you start thinking, “Well, what difference could I possibly make?” Right? Feynman could do stuff in his head that, you know, would take me a year to do.

Feynman wasn’t the best, by the way. The guy that I was most impressed with who I’m still most impressed with is Pauling. Pauling really, really had—Pauling was 20th century chemistry. And also a very unusual guy. I mean, this guy—one of the things I really took from Caltech is these were really unusual guys and we used to have them over. The way I got to know most of them—Feynman I got to know a wee bit differently—but most of them came to dinner at the house and they’d tell us their stories, right?

So I’ve heard Feynman’s stories about Los Alamos many times. In particular, many times when—what I remember is specifically he used to tell these stories that I’d heard them maybe two or three times by the end I was there, but his wife would usually accompany him, must have heard them thousands of times, right? And she would, of course, fall asleep while he was telling us all these great adventures. But, I mean, Feynman was incredible.

Pauling was incredible. What I remember about him especially is—so I was working, at the time, I joined the lab work around superconductivity and so I knew a little bit about superconductivity. And Pauling came over—and he was talking about stuff, he talked about nuclear physics, he talked about stuff that you wouldn’t expect him to talk about ‘cause he’s a chemist, right—and one of the things he said, “You know, nobody really understands superconductivity except me.” This is, of course, already after the BCS theory and stuff like that. And he started explaining this, what he called, resonating valence bond theory of superconductivity, right?

And, you know, so I heard this and OK, it was clever and stuff like that. I didn’t think that he really had a tremendous insight into superconductivity or at least about superconductivity that I knew at the time, but of course, was it 20 years later, there wasn’t even high Tc superconductivity then, the valence bond theory was resurrected by Fyl Anderson, right? Who actually, you know, gave credit to Pauling for it. So Pauling was really something.

Zierler:

You couldn’t limit him to biology. He was too smart.

Chaikin:

You couldn’t limit him to biology or chemistry, right.

Zierler:

What class did you actually take with Pauling?

Chaikin:

Zierler:

And what convinced you, as an undergraduate, to focus on solid state as opposed to all the exciting stuff that was happening in cosmology, theory, particle physics? What was it about solid state?

Chaikin:

Well, cosmology, there wasn’t that much yet because cosmology, if you—I mean, there were some things—but nothing like the explosion now because of the instrumentation that we have now, which really took us beyond. I mean, all that stuff was fascinating, but I actually liked working with my hands a lot, not just doing calculations and stuff. I worked with my hands. And what I really liked is the fact—even early on, you could do this in those days with hard condensed matter physics as well—you could decide, “Oh, I don’t understand something. I’ll go into the lab, I’ll set up an experiment, and by tomorrow I’ll know the answer.” You could do that and you did that.

And it’s a joy working with the experiment—first of all, it’s a joy working with your hands. And then seeing stuff that is amazing right before your eyes. You know, the next day after you design an experiment is just amazing.

The first time I saw super fluid helium, amazing, right? Look in the Dewar you cool down, you go pump it down (to lower the temperature) suddenly the surface becomes absolutely smooth. There’s no bubbling. It’s obvious that the heat transport from the boiling and stuff is now nothing the same. You see fountain effects, you see helium climbing over things. You see it with your eyes and here’s a quantum phenomena, completely a quantum phenomena that you’re just seeing with your eyes. Not splitting atoms, not from a cyclotron, not from particle tracks and stuff like that. You’re really seeing it with your eyes. And doing experiments like that were really cool.

And you had to do all sorts of stuff back in those days, you had to blow your own glassware. So we did experiments—there was an experiment that Feynman—I mean I talked with Feynman several times, but he wouldn’t know me, but he was very good buddies with my advisor, Jim Mercereau who invented the SQUID, the superconducting quantum interference device—and Feynman, even in his lectures, he talks about—in his lectures it’s just a water sprinkler where when you turn on the water sprinkler, it goes one way. And he poses the question, “What if you now put it in the fluid and you suck on it instead of blowing through it? Which way is it gonna go?” And he got interested in this problem. And Feynman was interested in that problem and suggested since there was a low temperature lab, he said, “Well, why don’t you do that with a super fluid and see what happens?”

So there was a little thing called the Kapitza spider, which was you flow by heating it up. You’d flow super fluid through it, and you could make it so that the arms come out like that, OK? Oh, you can’t see because of my (screen) background, but in any event, like that so it rotates with the arms. And so we blew one of those. That is we made the glass gadget and we made something be like that and sat it on a pivot and put it in superfluid and watched it go. That was pretty cool.

But afterwards, I took courses that weren’t required courses and I had some really good people. I always remember Doug Scalapino’s course and Brook Harris’s magnetism course and there were, you know, lots of really good people there who I got to respect and who many scientists who, you know, I still look back at their papers and what they’ve done since, what they did before and what they’ve done since.

So the first thing I worked on—well actually my thesis was on the Kondo effect in superconductors. It was based on the premise—which I think was sort of my own, but may not have been—which was that the best understood many-body problem at the time, maybe even now, was superconductivity. So why don’t we use superconductivity in order to study other many-body problems. So I think the title of my thesis is Probing Many-Body Problems With Superconductivity.

And so one of the easy ways of doing that was it was the Kondo effect, which was not yet solved at that time. It was known what it was. It was known that the resistance went up and why, Kondo’s work, but it wasn’t anywhere near solved and so it was lots of activity there. So a lot of my research was on trying to understand Kondo scattering, Kondo impurities in superconductors, which you could essentially see by several ways. One of them was by how much the transition temperature of the superconductor was reduced when you added the impurities, you could find out what the Kondo scattering was doing to the superconductivity. And I know Penn’s learned something about what the Kondo effect—the state that the Kondo effect was going through was.

And then while I was there, at Penn, I spent a year at San Diego. Originally I got sent out by my advisor. My advisor, Tony Jensen, was one of Bernd Matthias’ students. I don’t know if you know Bernd Matthias, but Bernd Matthias was a very well-known physicist, material scientist. At the time, he had three positions. One was at Bell Labs, one was at some national lab, Los Alamos I think, and the other was UCSD, San Diego. And he was my grand advisor, one of my grand advisors.

So my advisor sent me out to do a Kondo experiment, a simple susceptibility experiment on an apparatus that one of his fellow graduate students, Deiter Wohlleben, had at San Diego. And I got out to San Diego and I was supposed to be there for three months and I ended up spending the year there. Partly because—this was I think my second year at Penn or maybe my third year at Penn—partly because I discovered San Diego, actually there was some good science going down there and the weather was fantastic. Best weather I’ve seen any place in the world.

But really what happened is I did these susceptibility experiments. They were incredible because you’d have to get on top of this apparatus, climb on top of the magnet on top of the dewar, and sort of adjust the apparatus for every measurement that you did. So you’d go up and you’d adjust things with a wrench and you’d go down and you’d take the measurement and then you’d climb back up and you would adjust it and you could take the measurement, one of those things.

But then I went—somehow I made contact, I forget how—with Ted Mihalison who had a job at Gulf General Atomic which was down the road, and he had set up just what we really needed for the experiments on Kondo effect, he had set up an apparatus for doing low temperature tunneling experiments with a magnetic field, and so somehow we made contact with him and so I got supported as a consultant for General Atomic and I was one of, I think, three consultants that year.

One of them was John Bardeen and one of them was Hans Bethe, and those guys got, I think, $3,000 a day and I got, what I figured out—I got like 25 cents an hour.

Zierler:

That seems about right.

Chaikin:

It seems about right in terms of the contribution, absolutely. But anyway, I’d spent the year there and then that’s when I got involved in proximity effect tunneling and things like that. So that was actually very good because it broadened what I was doing from simple measurements, like resistive measurements, to susceptibility measurements and transport and tunneling measurements and things like that. Actually it was very good, it really broadened the experiments I could do and a lot of them I continued doing essentially until I went into soft matter. Well, essentially until ten years ago.

Zierler:

Out of all of this work, Paul, how did you settle on a dissertation topic?

Chaikin:

Well, as I say, I saw what was going on. At that time, Alan Heeger was somewhere in his Kondo effect days, he was still making progress on that, as was Tony Jensen. And I wanted to, you know—I don’t think I was actually given this particular project by them, but as I said, I got the idea that superconductivity was the best understood many-body problem. And so since they were interested in Kondo physics, we sort of used superconductivity to study the Kondo effect and that’s the way it came up.

And then when I went to San Diego to do these experiments, the first was a pure Kondo experiment, and then later we were doing superconductivity with Kondo and then it became tunneling in induced superconductors with Kondo impurities, and that’s sort of what the thesis developed into.

Zierler:

Do you remember who else was on your committee?

Chaikin:

Actually, I don’t. I can look it up. I mean, my advisor actually died before my thesis was done, so Alan Heeger is the one that signed my thesis.

Zierler:

What was it like to work with Alan?

Chaikin:

Alan was terrific. Even before my thesis was done, Alan had decided to go into this new area of organic conductors. He was influenced a lot by a postdoc that he had, Tony Garrito. Tony sort of brought the problem to Alan and Alan sort of developed it into some exciting, and controversial, science which is now a well-known field that stands on its own. But we didn’t know anything about it. I mean, Alan, of course, knew more than anybody after he started looking and he understood that there was really interesting, deep, good, many-body physics in it.

And so for I think a year and a half or maybe two years before my thesis, Alan organized the group to meet every week at night and educate ourselves in the physics that we needed in order to attack this problem, the problem of conducting organics. And in those days, conducting organics were conducting organic crystals. So they were molecular crystals, which had some slight conductivity depending on how you doped them and things like that or what materials you made.

Alan was terrific. It was really terrific because Alan let everybody talk. So he would suggest papers and you’d go back before the week came when you were to present and you’d research it and you’d figure it out and you’d explain to everybody in the room, I guess the group was maybe ten people, something like that, and you’d explain to everybody what you thought was going on and other people would chime in and say, “That’s not right,” or “This is the way you should look at it,” and the thing that was really, really good is there was no such a thing as a stupid question. Everybody really learned because you had the opportunity to ask anything. You wouldn’t assume that people understood everything and if you said something that people didn’t know, they would just stop you and you’d have to explain it.

So it was really good—both from learning from what other people did and from the way to do science—which is—it really should be there are no stupid questions—and the fact that you really tried to explain this so that everybody would understand it. And of course the only way you really learn anything is when you have to teach it. And so it’s a terrific experience.

Yeah. I actually remember one of the things I had to present was - somebody in Russia—Shchegolev—was sort of Heeger’s counterpart in Russia for a while, had done an experiment on thermoelectric power, the Seebeck effect. And I had never heard of this and nobody else in the group had, and the claim was that he was measuring some stuff that was interesting and so I think Alan assigned me—either Alan assigned me to give a report on that paper or somehow it came up and I decided to do it.

So, I never complained. Actually no place I’ve ever been has given me lots of money to set up a lab, but it’s always been—I’ve never hurt for funds to do what I wanna do, but it’s usually because I really like to make things. And so it’s typically, you know, I can live with what’s—if there’s something around, I can usually use it. I can build on it. My start-up funds at UCLA I think were 30 kilobucks and there’s enough to buy a couple nanovolt meters, a couple of function generators, a couple of lock-ins maybe and that’s it. I inherited a lab, though. There’s a lab from Hans Bummel who moved to Konstanz as, I think, maybe even head of the University of Konstanz, and he was a famous man for doing ultrasound experiments. And so I got some dewars and pumps and stuff like that.

One thing that worked, I remember the first year I was there, most of the time I spent fixing pumps, soldering together apparatuses, stuff like that on my own. Actually I really loved that. The problem with growing old, as I am now, or even growing older, is you just absolutely don’t get in the lab as much anymore. You feel guilty, or at least I do, feel guilty going to the lab because I figure I should be doing something else. I should be figuring things out, my job should be figuring things out, writing proposals, and doing stuff like that. You know, making sure my students are doing well, but I really, really love going to a lab.

Sometimes I’ll go into a lab—I’ll decide, “OK, I’m gonna do something,” and I will just go into the lab and I love working with my hands with tiny things, with microscopes, with, you know, binocular magnifiers or stuff like that, figuring out—as I say, you really have to understand physics even to figure out how to mount things. To realize what all the thermal conductivities are, what the impedance is, where your energy’s gonna go, where the heat links are gonna kill you, everything like that.

And so you have to do it all the time and you have to figure out clever ways of doing it and I love it. And so the first couple of years, when I was an assistant professor, I had built every single apparatus in the lab, I had built it myself. I soldered things together. I did leak checking, I did everything. And I got them to work and they were pieces of junk. They just barely would work ‘cause I didn’t care about how beautiful they were or whatever like that.

And then when I got it to work, I’d give it to a student thinking, “Oh, this student’s gonna make improvements. This is going to be much better next year or the year after.” And six years later I’d go back and the apparatus would still be working and nothing would have changed on it. They hadn’t done anything. And so I have racks and racks—I still have them in my lab—racks and racks of things that I welded and soldered and stuff like that and I absolutely used to love doing that. And I probably could do it now, I’m not sure, but I would consider it a waste of time and I would hate myself for doing it because I’d be having so much fun and if you say, “Jeez, I’m just having fun doing this,” I should be doing something else. I should be figuring something out.

Zierler:

And so Paul, in those young days at UCLA as you were setting yourself up, what were the big research projects you wanted to get involved in? As you emphasized, what was most fun to you at that point? Was it thermo power first or that came later on?

Chaikin:

That came—well, I think I started doing it immediately when I went there, but I was doing superconductivity also a lot. I mean, this idea of using superconductivity to study other phenomena, I thought, was a really good idea. So for example, I mean one of the reasons I got this idea is that I always thought the real proof of the theory of superconductivity at BCS were these experiments that I think—just a sec—anyway, these tunneling experiments where they could unravel the phonon spectrum and you could measure the phonon spectrum then better by doing superconducting tunneling than you could by neutron scattering.

And the fact that from the spectrum that you calculated from it, you could feed back into the Eliasberg equations and exactly get things to work was essentially, I thought, the best proof that the theory of superconductivity was right. And it also gave you the best information about the phonon spectrum and the electron phonon interaction and other materials properties. And so I thought, “That’s really cool. Why don’t we take normal materials and induce them superconducting and study their electron phonon’s interactions.” And so that led to lots of work that we did on the proximity effect. And so that was another big thing that I was doing, but it all was under this rubric of using superconductivity to study many-body of things.

But the thermo power was also really neat because—so UCLA was a terrific experience. , Fyl Pincus was there—the real reason I got a job at UCLA is that Fyl Pincus and Alan Heeger had always been buddies. They’re still buddies. As a matter of fact, I think Heeger lives in the apartment above Pincus’s now in Santa Barbara. But they were buddies since Berkeley, since they were graduate students at Berkeley. And so Fyl is really the one that got me to go to UCLA. And Fyl is a little bit nuts, but he’s a great guy to talk physics with and he knows a lot and stuff like that, and he sort of wouldn’t let me get away without—he’s a theorist—and he wouldn’t let me get away without doing some theory.

And so this was terrific because I learned that you don’t have to—I shouldn’t say this—you don’t have to be a genius to do theory. It’s just a matter of having an idea and doing the best you can to figure out where it leads you. And so when I got to UCLA—no actually, sorry, I’m wrong. Actually I did theory with Alan before that.

One of the things I did—I did a couple of papers when I was at Penn with Heeger on theory. I forget who else, but in any event, it was on excitonic polarons and the guy—so that’s another reason I ended up at UCLA—the real hero of polarons was Ted Holstein, who you may or may not know, but he formulated the Holstein-Primakoff transformation, but essentially he was sort of the senior theorist at UCLA. And his work on polarons inspired what Alan and I had done on excitonic polarons when I was at Penn. And even Fyl, and a little bit Ted Holstein, sort of gave me this impression that it was OK to do theory, right? That you could do stuff like that.

And so I did some stuff with Fyl’s graduate students on thermo power and those papers have actually lasted, they’re cool and still cited. Anyway, so UCLA was really good because it was really an amazing group. All the condensed matter people were on the same floor in the same corridor. And Holstein would, every morning, go from one—work his way down the hall, go into everybody’s office, ask what you were doing, and then let you know whether he thought it was worthwhile. Whether he thought it was complete nonsense, or whether what you were doing was interesting. So it was really terrific.

And Fyl, of course, had already almost made the change when I got there. He was still doing some electronic physics and still interested in it, but Fyl was the disciple of Pierre Gilles de Gennes. He and de Gennes were big buddies. And so Fyl, I think, is the guy who really brought soft matter to the US from France, from Europe, and Fyl tried for 10 years to get me interested in soft matter. And I didn’t budge. I just didn’t do anything with it for sort of the first ten years.

But nonetheless, I had a great time at UCLA and I learned a lot and it was a terrific time.

Zierler:

Did you continue on with the Kondo effect? What was the status of that field by the time you had gotten to UCLA?

Chaikin:

By the time I’d gotten to UCLA, well, it hadn’t yet been solved, but it was being solved.

Zierler:

Right.

Chaikin:

By Wilson and Anderson. So it was still interesting, but I mean—and there’s lots of work that was done after that—but somehow I wasn’t—I don’t remember why I wasn’t particularly interested on it anymore. It seemed to me that this was a field—so I got more involved—most of the time I was at UCLA, the electronic physics I did was on the organics. And if you like, the reason that I found that much more interesting is this was a field that was absolutely just opening up. It wasn’t like, “OK, here’s the Kondo effect. We have 10 years’ worth of research and we understand this and we understand that and here is actually a theoretical solution to it. We now know what it’s gonna do.” It was, “Well, we have all this vast class of new materials, they’re showing different effects.” This is when charge density waves first and then spin density waves were coming in. There was always the possibility of them being superconducting, but they weren’t.

And the field actually rested on this idea by Bill Little, which never worked out, that the organics may give you excitonic superconductivity at room temperature and above. So that’s what propelled a lot of the field and the science was fantastic on its own with the discovery of new effects.

So I think it was the fact that it was a new field and there were actually completely new things to be discovered that took me away from Kondo and maybe even superconductivity to the organic conductors.

Zierler:

And what were some of the technical advances in organic conductors that you saw really providing a lot of opportunity to move the field forward?

Chaikin:

Well, I mean, part of it, of course, was the chemistry and part of it, where almost everything in condensed matter, was new materials, right? When you discover you have a new set of materials, you have a new playground. And unless something is wrong there, there has to be some really interesting new physics almost anytime you get new materials.

So there was that. There was the observation of new phenomena, and the fact that a lot of it was new physics in the sense that there were lots of nonlinear effects and there were things which were not equilibrium effects that came in. But I think there were all sorts of questions that resonated with new models and new physics, with dynamical models, with questions of dimensionality and incommensurability with the density waves, with the dynamics that you get from nonlinear effects and couplings. So I think that’s what made it really interesting.

And, of course, then, the idea—the thing that really made the field blossom much, much more so than it had before—was the discovery of conducting polymers, right? Because the conducting crystals were essentially really, really interesting physics, but nobody ever imagined that they would be applicable practically, that they would be of any industrial use whatsoever because they were fragile, they were small, they were temperamental. You never imagined that you could process them easily. At one point they weren’t even superconducting, then they became superconducting and interest rose again.

And that showed lots of things and the fact that you could do Landau theory on the phase transitions that you see in crystals. Those are real insights. And the fact that when you—and of course it’s realized now, you know, some of this stuff which Landau did became the Phi four field theory, which influenced all of physics. The insights that you have from constructing order parameters for soft matter systems or phase separation, which is essentially now part of soft matter as well.

So I think the field really—and I may be—that’s my impression, that it came from a couple of people who were card-carrying physicists realizing that here is a field where you could really not only have impact on the field, but have the field have impact on Physics. So deGennes sort of did superconductivity for a dozen years or so, did magnetism, did superconductivity, did fabulous work in it. And then he said, “I’m not doing this anymore, I’m doing liquid crystals. I’m doing polymers.” And started making contributions in those areas and a lot of it was mapped directly onto what he was doing in superconductors and phase transitions and magnetism and things like that. So that’s my viewpoint of where the field came from and—

Zierler:

And then specifically on you, where did you get involved?

Chaikin:

So what happened is—like I say, Pincus had been trying to get me for years to do it and I never really thought it was that great or that interesting, stuff like that. So I took a sabbatical at Orsay in Paris and I was supposed to work with two groups, one on superconductivity on proximity effects of superconductors, and the other on organic conductors. But my host was Etienne Guyon who was one of deGennes buddies and disciples in Paris dating back to when they were at Ecole Normale, and he put me in a room with his graduate student Pavel Pieranski.

And Pavel is an amazing guy. Everything he does is not only beautiful physics, but beautiful, and he walks in one day to the office, and he says, “Regardez ca”, “Take a look at this.” And I look at this little vial and it looks like liquid opal. And I said, “What’s that?” He said, “Oh these are colloids, these were charged colloids, they are charged and they form a Wigner crystal, they’re crystalline, and what you’re seeing is the Bragg scattering.” I said, “What?” And I took it and I just played with it and it was just amazing because what you could do is discover things by shaking and watching the colors disappear and oscillating to see the resonant modes in the cylindrical bottle—and this was just within 10 minutes—he was showing me stuff and I discovered stuff on my own—you could essentially twist it around your hand and you could melt it. So you could see the difference between the liquid and the crystalline phase immediately.

I mean, I think that we need collaboration with biologists. I think it’s probably good for them. If you mean having biologists come in and take over the field, I think no. And I think that soft matter is different from biophysics. I think biophysics, at least the way it is now—and this probably will get me in lots of heat—is mostly using physics to explain biological phenomena.

Zierler:

You mean as opposed to the other way around?

Chaikin:

Well, as opposed to—I think what soft matter does is it essentially asks, “Well, OK. What phenomena do we have with complex systems? How can we understand a phenomena or the physics of what goes on rather than to explain why this works, rather than figure out why, let’s say, microtubules mechanically grow and then explode, can disassemble themselves immediately.” So figuring out what the physics is in the molecular assembly, specific molecular assembly for those molecules coming together.

If you find in general what causes structures to be stable or unstable, you’re doing soft matter. If what you do instead is you say, “OK, I’m gonna figure out how a motor protein works by essentially looking at how the ATP is used in a molecular motor to get the action,” the mechanics of that, the chemistry of that, I would say—this is really gonna get me in trouble—I say that’s biophysics.

If I want to know something like how, in general, do I imagine molecular machines working? What’s the overall operating principle by which they work, I’d say that’s soft matter. So between those, for example, is one of the examples I always like is the ratchet model, the athermal ratchet model that Jacques Prost made many years ago for trying to describe the commonality of molecular motors. That, I’d say, is soft matter, whereas if you’re looking particularly at how myosin works, it’s not. Seeing what happens in general when molecular motors assemble or what they can do or how they assemble in general, I’d say that’s soft matter.

An individual one is biophysics. Wow, this is really gonna get me in trouble. OK. [Laughs]

Zierler:

Please.

Chaikin:

You want me to get myself more in trouble?

Zierler:

Absolutely.

Chaikin:

Well, I don’t know what else I can tell you. I mean, biologists— to me—have two tasks. One is to actually figure out how biology works. They don’t have to make a contribution to physics, right? They have to figure out how biology works and in some sense, their responsibility is how it works microscopically and on another scale, you know, how it works globally.

How it works as what are principles of organization of biological systems, which are specific to biology. And I think—I hope—that the condensed matter of physics can also do some of that, but in general, condensed matter physics is trying to figure out how things organize on a macroscopic scale from the microscopic scale. They’re trying to figure out condensed matter, soft matter is about how that operates not only for biology, but in general. Biology is part of it. I don’t like the idea of mimicking life so much because—although everybody tells you that biology is optimal—what I’ve never seen in biology—what I always say is I’ve never seen anything in biology make an I-beam. And an I-beam is structurally much better than anything else for what it does. So it’s not that optimal is the answer. Actually my opinion has always been all biology has to do is work. It really is not required to do anything optimally or otherwise. If it does, that’s fine, but I don’t know whether that’s a general principle.

But I’ll leave that to the biologists. I think that we can really impact on that, but the role that soft matter has or condensed matter has in general is figuring out cooperativity and maybe using it as well ‘cause a lot of things that we do are useful. A lot of the things that we discover about cooperative things, how you build things, how you assemble things, what you can make and what you can’t make and how to activate them, that’s all in the purview of condensed matter and especially soft condensed matter.

Zierler:

Paul, let’s talk a little bit about your move back to the east coast. What was the circumstances? Did you accept a joint position at Penn and Exxon or did they happen separately? How did that work out?

Chaikin:

It worked out in a funny way. What I thought—I was at UCLA. I was fine, I liked it there and everything, and so I looked at Penn and I looked at Exxon and I said, you know, they’re both really good. They’re both essentially each individually not necessarily better than what I have at UCLA, but if they were combined, it would be definitely better. [Of course the number of people, the contacts and stuff I would have with really good people that I really respected and stuff like that would be higher.

And so I told people at Penn and Exxon this, and they were going to try and give me a joint appointment, figure out how to do a joint appointment. And I was guaranteed that they could do it, well of course they couldn’t do it. It turns out when they finally got together to try and do it, they said, “There’s no way that the intellectual property da, da, da, da, can possibly by handled. There’s no way we can do this.” So what they arranged is that I would be a non-regular employee at Exxon at no more than 49% and I would have a regular position at Penn. And so that’s what it turned out to be. That was the arrangement.

Chaikin:

So he is the father of that field. And he was at USC for a long time and when we were together, we collaborated on several papers and we fought on others. We were at odds on other papers, but when he went to Princeton, and some years after he went there, he contacted me and said, “OK, what about coming to Princeton?” And so I went up and saw what they were doing and what they were gonna do and stuff like that and it was closer to Exxon. There were some really good scientists there, of which you certainly know some.

Zierler:

And how strong was Princeton at that point in soft matter?

Chaikin:

In physics, it was zero. Even more zero than it is now. Well, I could also take it back. There were already two biophysicists there in the physics department. Bob Austin and Saul Gruner were there as biophysicists. They didn’t have any soft matter physics, but they have a good chemical engineering department and I had known or I knew of—from their research—several people in chemical engineering at Princeton.

Yes. So what happened is, I mean, there’s this problem people have studied for literally thousands of years - the packing of spheres. It’s an ancient problem and it’s been solved by grocers because when you go to a fruit stand, you see that the oranges are stacked up in a lattice and it turns out it’s a Face Centered Cubic lattice and people have known that and stacked them that way forever, since there was fruit, and it’s been—this thing is also known as the Kepler conjecture because Kepler raised the problem of the densest packing of spheres, and again, Kepler conjectured that it was this FCC packing, cubic packing, but it’s only been proven mathematically that that’s the densest packing maybe 15 years ago. So mathematically, we now know that’s the densest packing.

But it turns out that if you just take spheres and you pour them in a container, they don’t make a crystal. They make something that’s disordered, which is like what happens if you go to your grocer and they don’t stack them up, then you just see the oranges all over the place. And it’s known and it’s been known for a very long time that that doesn’t pack as well and instead of going to 74%, which is crystal packing, it packs to something like 64%.

Zierler:

74% means that there is 26% empty space?

Chaikin:

Yes. So for example, if you fill a box with marbles and you put them exactly in an FCC lattice and then you pour in water and you drain it out, the water that’ll come out is 26% of the volume of the box, just as you said. OK. So if you do it randomly, that is you just pour them in so they go everywhere and they don’t form this crystal, they fill space to 64%, which is less.

So it turns out that problem—everybody sort of knows that answer, but nobody has done anything in order to prove or even show mathematically or even define the problem mathematically. And the reason is because nobody knows really what random means. That’s really why it’s a real hard problem. People think it’s an undefined problem.

On the other hand, they know something about it. So one of the things—this is kinda cool so I’ll tell you, but I’ll make it short ‘cause I could go on forever on this. So Maxwell made the following observations; suppose I know nothing else about its system except that it’s stable. Except that nothing is moving in it, it’s sitting there and it’s stable. I can still say something about it. Suppose I have particles there and these particles can have any arrangement you want with them. But nothing is moving so everything’s stable. So he learned something just from the fact that it’s stable.

What I have to do to have it stable is that I take a particle and I know that it can move in three—let’s say it’s in three dimensions—it can move in three dimensions. That means I have to specify—for it to be stable—the three coordinates, X, Y, and Z for this particle, right? OK. So I have three numbers for that particle, three variables. Now if I have N of them, I have lots of them, N together, that means I have 3N variables. And for this to be stable, I have to specify them.

And we learned in public school that the number of equations has to equal the number of variables if you wanna solve it, right? So I need 3N equations. But if all I know is that these guys are particles and they could be touching or not, the only way I can get an equation is from a contact. So I know that there must be three N contacts. And since a contact has to have two neighbors, this is a neighbor of that and that’s a neighbor of this, so that means on the average, every particle in this random pile has to have six neighbors, which is really clever, right? And that’s from Maxwell.

And the idea was if you’ve got something that’s not, you could have more than that, more neighbors but you can’t have less than that. You can over-determine the problem, you can under-determine. So people have studied this problem and indeed, people have found that when you have marbles randomly thrown in a box, there’s six contacts, six neighbors, and the packing fraction is .64 and that’s all that anybody had done for thousands of years. So I said, “Well, what if you use a different shape?” Some reason I got in my head what if you use a different shape? What’s gonna happen? OK. And you know for instance if you choose cubes, you can completely fill space. Of course if you pour them in randomly, you don’t know what they’re gonna do. What you can do is you can just pack face-to-face and it’ll completely fill the space 100%.

So I said, “Well, what’s the simplest object other than the sphere?” Physicists always use spheres. It’s the simplest object, right? So what’s a small perturbation on a sphere? The answer, it’s an ellipsoid. OK. So why don’t we try packing ellipsoids? So now the question is where do you get lots and lots of ellipsoids all the same shape and all the same size? Answer; M&Ms. OK? So—

Zierler:

Now Paul, did you have an amazing, comic book moment where you were eating M&Ms and you looked at it and you said, “Ah-ha”?

Chaikin:

No, it was actually was more like what I told you. It turns out though everybody thinks that that’s the case. Everybody thinks that I got it that way because for many years before this, my lunch was always coffee and M&Ms. Everybody at Exxon knows that, everywhere I go to lunch, the only thing I would have for lunch is one bag of M&Ms and coffee. So everybody thinks that that’s it, but in fact it wasn’t that. It wasn’t an “ah-ha” moment, it was, “Well, what can I use?” And I was thinking maybe I can flatten some spheres and I can make some, da, da, da, and I was looking around and it’s like, “Oh yeah. I can use M&Ms.”

And then I went and I had somebody measure and it turns out they’re all the same size and they’re all the same shape to better than half a percent and we did all sorts of studies on it. And then I got somebody to do the experiment. I got an undergraduate to do the experiment. And then—so I got one guy to do the experiment, he did it, and he found that they pack better. It packs more densely. It packs essentially to 70% instead of 64%.

So then I had what I considered this lightning moment, which was OK, the interesting thing is why does it pack better? I mean, first I didn’t believe it. I told—I had a student do it and then I had another student check it and then I did it myself because I said, “What the hell?” I know that I can get from a sphere to an ellipsoid just by squashing.

So if I take a box full of spheres and I squash it and nothing changes, the packing’s gonna be the same because the volume—well, anyway, the volume of an ellipsoid is 4/3 pi abc, instead of 4/3 pi r3. In any event, I get the same volume fraction – filling factor. So I didn’t believe it and then I had a couple more students to get them to do it and then? Why better? And that’s the time I had the Eureka moment, if you like, which is I said, “Oh, it’s because when I have an ellipse - if I have a sphere, all I have to tell you is X, Y, and Z. But if I have an ellipse, especially and ellipsoid of revolution, there’s a direction along the axis of rotation and that has two angles associated with it and that means that the number of degrees of freedom I have to specify is XYZ and two angles.

So instead of having three equations per particle, I need five equations per particle and therefore instead of having six neighbors, I have to have ten neighbors. And if I need ten neighbors, I have to pack it denser than if I only have six. Does that make sense?

Zierler:

Yes.

Chaikin:

So then I got a summer student who was from another university to sit there, pour paint in, drain the paint out, dry it, take the things apart, and see where—on the M&Ms—how many patches were there with holes in the middle. So this guy spent the summer doing that. And there should be—instead of for the spheres he counted—and as you expect, there were six—that’s also what other people had counting and with the M&Ms, it was 9.86. So essentially ten. So that was really, really, really cool.

And so there’s a picture of me—I think there is with Sal Torquato, somewhere standing with Sal who’s someone who used to be a chemical engineer—he’s also a chemist and a physicist now—in front of a big barrel of M&Ms. And so that’s what—probably actually if you just Google my name, that’s one thing that’ll come up. So that’s essentially what I’m known for.

Zierler:

Well—

Chaikin:

And it’s actually kinda cool. And the thing to me that was great, I mean I really—when I figured this out—actually when I figured it out, I knew that that was the answer and the experiment was just the thing at the end, but the thing that got me is I just went around for weeks after that thinking, “Jeez, I just have the best job in the world. This is just the best job in the world. I get paid for figuring out that the reason that ellipsoids pack better than spheres is that they’ve got 2 more degrees of freedom. It’s just amazing that people pay me for this.” So I was just ecstatic with this.

Zierler:

Were there commercial applications for this? I mean, are people in the packing and shipping industry, are they interested in the significance of this finding?

Chaikin:

So far, I think the answer is no. And the people that actually should be interested in it are people that make ceramics. The reason is with a ceramic, you know, are particles in a solution and they mix them around and they put them in a container and they mold them and the strength of the ceramic—comes from how many contacts you have and the density.

And it turns out if you use spherical particles or something like spherical particles, the best you can do is 74% and the number of contact—actually the best you can do is 64% if you just pour, and the number of contacts you get is 6. Whereas if you take ellipsoids, not like M&Ms but with three different axis, the best you can do is like 77% and the number of contacts you get is twice as many, it’s 12. So you should be able to make much better ceramics that way.

Now, the interesting thing though is that the candy company which makes M&Ms, they didn’t know that M&Ms pack better, and of course that’s exactly what they don’t want, right? Because if they pack better, that means if you sell a bag full of M&Ms, you have to put more of ‘em in to fill it than if they were spheres. So in fact, what would be the best for [the] Mars [Company] is having spheres in there or something even worse that would pack less densely. Then you can sell for the same price fewer M&Ms.

Well, they called me up for a recommendation for somebody, which I gave them, and I said, “But you’re in New York, you should take advantage of it.” New York is the best place in the world. Anyway, so I told them that and so I get a call back and they say, “Listen, why don’t you serve on an advisory committee?” They said, “We have lots of retirements coming up, we haven’t hired people for a long time. Why don’t you come and serve on our advisory committee for another year, couple of years, and tell us what you’re gonna do?” Or, “What we should do,” I should say. “Tell us what we should do.”

So I joined it and I came for a couple years and they had some plan where in fact they had—they were gonna have 12 positions open to fill in the next four years or something like that, and so I listened to their plans, I met with lots of people, and their plans was gonna be, “Well, we’re gonna replenish this group or we’re gonna replenish that group and we’re gonna replenish this group, maybe we’ll start something over here.”

And I just said, “No, this is just ridiculous.” I said, “If you wanna build something up, what you wanna do is you wanna build something up which uses your positions in a smart way and makes you instantly, or in a very short time, one of the top departments in the world, or the top department in the world in this area. And you don’t do it like that. The way you do it is you take your 12 positions and you pick one or two fields and you just put all your positions in those fields.”

And so for a year or two, this thing went around and they said, “Well, what areas are open?” And so I said, “Well, you should do—“ I forget what I said. I think something like, “You should decide. You should either do quantum computing or soft matter. Something like that. Those are fields that are opening up. There is nobody that has the dominant thing in that. There are groups everywhere, but you should take advantage of it and do one of those two fields. It’s up to you which you wanna do.”

So after a while, they said, “OK. Why don’t we give eight people in soft matter and you do it.” So, eight people is a lot of people. And I talked around with some people and I told them I could get other people to come and I talked with Dave Pine and I called him and I said, “So when are we gonna form a soft matter group at NYU?”

So sort of from that time on—that was just the year before or the year that I came to NYU—I sort of decided what’s really interesting is the dynamics. Some of the dynamics we solved. You’re not gonna learn anything new from thermodynamics. You may see a couple new phenomena and stuff like that, but most of the new stuff is nonequilibrium, had to do with well, how does nature do stuff? What’s the physics of life, if you like.

So we started working—from that point on, essentially since I came to NYU, most of the work has been on active matter of one form or another. I have a group with Ned Seeman here who’s in chemistry. The guys I work with are great. All the guys I’ve worked with through the years are infinitely better than me. All the guys I’ve worked with have really done something, like Bill Russell, for example, he’s really amazing and stuff like that.

Ned Seeman, who I work with now at NYU, is the father of the field of DNA nanotechnology. He is the first guy—I don’t know if you know him—but he is the first guy that ever came up with the idea that, “Oh, you know this DNA which recognizes stuff and everything like that, you can actually use that as a material and build things with it.” He founded this field “DNA Nanotechnology” with maybe 300 institutions in the world now that work on it and stuff. Anyway, so I collaborate with him. He’s fantastic.

And so first the idea was we wanna take particles that recognize one another and so this had been done before, but we were the first to do it with colloids and Pine was part of this and other people were part of this. So we just took colloids and put DNA on them. And now you can make it so that this colloid will only stick to that one or only stick to that one or stick only to these two or whatever. And now we’ve designed all sorts of things where you can put the colloids together.

And then the real thing I wanted to do because of this list that I made, that Tom and I made for Dave Weitz, is to make an artificial system which would exponentially grow, which would replicate, which could exponentially grow and could mutate, right? So I wanted something that could show that we knew enough about this to actually design a system that could do elementary life, essentially. If you try and think of what life is, nobody of course knows how to define what life is, but one of the components that usually people associate with it is self-replication, exponential growth, the ability to evolve.

So we did it. We have a system that does that. It’s made of tiles that are made from DNA, but they don’t use the DNA replication. They just use the DNA so that they can attach to a specific other tile and they do that. So most of the stuff we’ve done since I came here is active matter. We did all sorts of stuff with active matter, like matter that you shear. We made particles which are swimmers and the swimmers swim, which means motile, they have a velocity going forward. When you turn on light they swim and they stop swimming when you turn the light off, and they do all sorts of interesting—show all sorts of phenomena.

We have particles which we can spin and those guys are really interesting. They’re the first example ever of a system that self-assemble from hydrodynamic interactions alone. So the only interaction between these particles is through the fluid. Essentially you have a particle that’s rotating that creates a flow field that’s rotating and that advects other particles and it turns out they set up a flow field so that they clump together and roll as what we call a “critter”. And that’s the first time that people have ever assembled things just from their flow field.

You’re not gonna solve that, right? You’re not gonna solve that and close the book on it ever, I don’t think, right? But I got pushed by NASA to hold—so I ran a workshop 17 years ago on what was hot in soft matter, essentially, right? So we ran a workshop in March on “grand challenges in soft matter”. And I just wrote sort of the first little synopsis of it. So I can tell you what I decided from that, which is taking together the ideas from everybody in this workshop. There were maybe 130 people that were in this workshop, which was all Zoomed. We scheduled this for the March meeting but the March meeting was canceled, so we had it on Zoom.

But essentially what comes from it, which isn’t just my ideas, but sort of by reading of community is indeed nonequilibrium phenomena are what we should be working on in the 21st century. We want to know the organizational principles for things that are living or life-like and how to make them. So we want to know how things that are life-like work. What do I mean by that? So one of the big things is suppose I have—what we’re used to in soft condensed matter is how do you self-assemble things? You take things and you have forces acting between them and you can self-assemble them and you can make them in different shapes and stuff like that. OK. But that’s old. Don’t tell everybody that that’s old because there are a lot of people that are still working on that.

But suppose instead of doing that, you make a little machine. So instead of using a colloid, you take a colloid, which also swims, which also recognizes another colloid because they fit together or because they do this, that, where you can turn the swimming on and off, where you can make this essentially a little machine. So you take a little machine and now, instead of using your individual little colloids to make something, to assemble something, you assemble it out of machines. So you have a machine that’s made of machines.

Well, as you’ll immediately see, everything that’s alive is a machine that’s made out of machines. So this is sort of asking the general question, how do you do that? How do you take something which has active particles in it, particles that can sense, that can change, that can move, have them interact, and figure out what are the problems associating with that system? And it’s a nonlinear system. You have feedback in that system. It’s a complex system. And it’s a fundamental question in physics, how do you take a dynamical system and will it always go chaotic? Is there any way of controlling it? How will it organize? You know, these are basic fundamental problems.

And so I think that for example, nonequilibrium phenomena, dynamic self-assembly, something which has an amazing amount of complexity in it, how does that work? How can you do that? How can you figure something like that? How do you figure out its stability? And it’s sort of a fundamental question on life, on biology, but it’s also very physicsy, a question on the essence of biology, right? How do you get something like that?

Now you can also ask, on a different scale—but it also sounds very related—how can you make an ecosystem? In other words, how can you build something where although they’re still not connected together, you know, like a machine, like building something like that, but you have lots of pieces that are, themselves, acting cooperatively and now you have something else is that—how can you get those to be sort of an ecological system so that they feed back off one another? It’s another level of complexity - stuff like that and it’s a problem that’s a general problem in society as well, or for instance for anything you’re doing or for any way that societies or biological systems or active systems in general are gonna interact.

So that’s one of our grand challenges. Grand challenge? Let me see for a second. I forgot what I put down for this. I mean one of the grand challenges, although it’s not written as such, is artificial life, but that’s something else. So the third one that we came up with or that I tried to distill from this workshop is active meta-materials. That is to say you can tell all of this has to do nonlinearity, activity, and being out of equilibrium, but meta materials, you know, that have been made so far are static. They are inanimate and things like that.

And so the idea here is you can make, for instance, fluids where the fluid element—the particle the fluid is made of—has its own energy source and its own activity and hence, it’s not something where you act on it and you dissipate your energy in it, but it has its own behavior and it can give energy out to a system rather than simply absorb it. And making more and more sophisticated machines like that.

One of the things that we wrote here is that, you know, it looks like—in this field—especially since becoming more complex and dynamic and stuff like that—things like machine learning and stuff are gonna be—are starting to play a role and they’ll play a bigger role. And it’s also probably that—I mean, so people are starting not only to use machine learning essentially to analyze data and stuff like that, they’re using it to do experiments.

So you do your machine learning with the experiment as well, right? And that’s a new way of doing experiments. So you let the program determine which way the experiment goes and have it converge or diverge or whatever. And of course in doing this, when you do the experiment, you do that—you are also learning something about how machine learning works and feeding back on it. But indeed, we expect that there’s gonna be stuff feeding back.

There are people—I’m not one of them yet—who are doing things like—you can make robots. They’re not quite the size that we wanna do, which is micron scale, tens of micron scale, they’re on millimeter scale, little robots that signal to one another and you can just put them down and turn them on and actually have them move to their power source and have them do things. Like for instance, since they’re powered by light, they wanna go toward light.

And so you can have a simple program and a chip in them that wants to find light. But now you have many of them that are fighting for the same light. And now you can also put in them that—and this is not science fiction, this is of course real experiments that people are doing right now—you can put little LEDs on them so they signal to one another. And now you see what they do cooperatively, and typically a system with even those few features is complex enough that you have no way of predicting what they’re gonna do. You essentially have to do the experiment to find out or put the program in with machine-learning and have it feedback and see what it does. But as soon as you get something that’s even that simple, yet complex, the problems are more than we know how to handle. So there is a lot to be done. There’s a lot to be figured out. Just knowing the forces are not enough.

Zierler:

On that note Paul, there’s a lot to stay tuned for. I would like to thank you for spending this time with me, it’s been quite a ride listening to all of your stories and experiences throughout the decades so I really appreciate it and this is gonna be quite a special edition to our collection. So thank you very much.

Chaikin:

OK.