Steven Chu (2015) - A Random Walk in Science

Thank you very much. I just want to proceed. I first want to start, this is not going to be a usual talk. It'll be more a quasi-autobiography to emphasise the fact that life in general is a random walk. I want to show a picture of my parents - a very handsome couple, getting married in 1945. They came from China to go to graduate school here in the United States at MIT. You'd think that a couple like this, how could they have a child like that. (Laughter) That's me. It's a regression to the norm I think. In any case, I went to the University of California Berkeley and I spent eight years there as a graduate student and a postdoc. I want to tell you a little bit about the things I worked on. For example the first project assigned to me by my thesis advisor was a theoretical project. I was a mathematician and a physicist when I was an undergraduate and thought I was going to do theoretical physics. They said that was fine, I could do theoretical physics, and that was the project. I worked on it for three months and decided I enjoyed going in the laboratories, so this is an incomplete. I'm giving myself grades. The next thing was, I love music and I'd listen to a violin play music very rapidly, a passage. If the violin has missed a note you could pick it out. I did a little calculation and said, the uncertainty in frequency and the uncertainty in time has to be at least equal to one or greater than one." I set up a little experiment and had my subjects, fellow graduate students. I would tune a frequency, they would tell me by tuning another dial what frequency it was. The good people could violate this uncertainty by about twenty-fold, twenty times less than delta ny delta t = 1. I spent a week on that, it was a success and that convinced me; it wasn't published but it convinced me I really should be an experimentalist. Then I worked beta-decay of weak interactions, I spent a year on that. My advisors and I both said, "Let's look for something more interesting." So I got another incomplete. I worked on using lasers to excite highly ionized hydrogen atoms in an accelerator to measure the Lamb Shift. I spent a year on that, another incomplete. You have to understand, by this time I was a graduate student, I was three and a half years into my graduate studies. I have a list on incompletes, one success - never published - and about to begin another thesis experiment. This is really an auspicious beginning. What happened next? What happened next is both my advisor and I felt that there was a potential using atomic physics to test a theory that unified weak and electromagnetic interactions. I said, "I want to do this experiment". He says, "I agree." We stopped what we were doing and we started to do this work. I worked on it my final two and half years as a postdoc. I give myself a C on this because we didn't measure what we wanted to measure. It was much harder than we thought, so my first scientific paper was published a year after my graduate studies. It said, "What we observed is highly forbidden transition, but we didn't observe the parity nonconservation effect." I told my advisor, "I've suffered enough as a graduate student, why don't you give me a PhD? I'll stay and work on the project." And he said, "Fine." So I stayed another two years and that lead to this paper, a preliminary observation of parity nonconservation. You'll notice something: this is my thesis project, this was my postdoc project and I'm one, two, I'm third author on this. I didn't really care about the authorship. Why didn't I care about it? Because by six months before this paper came out, the Berkeley Physics Department made me an offer to become an assistant professor even though I only published one scientific paper. I accepted and I took their startup money and that was that. Having one scientific publication and a C minus or a C scientific publication, they actually offered me a job at what then was the best physics department in the country. Somewhat unusual, so unusual that they said, They didn't know my failures. They said, "You can take the job, start your group now or you can go somewhere else, anywhere you want, but the job is yours." They put me on the catalogue and I went to Bell Labs. That's me, with more hair and looking a little younger, adjusting a laser I designed and built that we use to do this work. It's a dye laser. With my start-up money, I used it to buy a CW dye laser. That CW dye laser is over here. This is a much better laser for a future follow-on experiment. The first year graduate student working with me on this project was Persis Drell. We machined the transition stages and things like that all in the machine shop. It was a wonderful experience for both of us. She then became the director of SLAC, Stanford Linear Accelerator, and is now Dean of Engineering at Stanford. In any case let me continue, I took my two year leave of absence to go to Bell Laboratories. It was a truly wonderful place. I bought a house right here, and would hop over the fence and walk to work and spend a lot of time there. Hop over the fence and go home. In Bell Laboratories - I have to also point out, that it was a remarkable place, they hired all their scientists very, very young. Fresh out of PhD or just completed postdoc like I just did. Of those, fifteen of them went on to earn Nobel prizes. My life at Bell Laboratories had ups and downs. The very first experiment I worked on was an experiment in energy transfer in a substance called Ruby, Al2O3. You excite a chromium ion, it transfers energy to another one and another one and another one. It was used to test a theory called Anderson Localization. Anderson Localization was this prominent theory invented by Phil Anderson. He had gotten a Nobel Prize one year before that and he said, in his speech that the studies of energy transfer in Ruby were the only clear demonstration of Anderson Localization in physics. I thought with Hyatt Gibbs and Sam McCall, that was done with a Ruby laser that can only tune to one side of the absorption line. We can do a better experiment with a dye laser and really nail down that this is in fact a clear demonstration of Anderson Localization. What we found instead was we couldn't reproduce the results, went on to measure how the mechanism of energy transfer actually occurred in Ruby and found it was a long range dipole-dipole interaction and Anderson Localization didn't even apply. That was what you would call a failure. It's a non-result but it actually led to my first tenure offer, amazingly. Then later, or concurrently, I did some spectroscopy on an atom, an electron and its antiparticle, clearly in the business centre of AT&T, with Allen Mills and Jan Hall. We did a series of experiments with successively better and better spectroscopy but ultimately we measured the 1S to 2S level in this atom to four parts in a billion, but it didn't lead to anything new. It was considered a technical achievement but it really led to no new discoveries. There were a few other experiments I did, a couple of incompletes. But I started to work on laser cooling and trapping of atoms and that actually worked better than I expected. Let me briefly explain what this is about. This is a charged rod. It's charged positively. If you hold it next to a piece of paper the electric field induces the charges on that piece of paper to move around. And so you see that the negative charge is a little bit closer on average than the positive charges. And since the electric field is higher in the region of the negative charge, the field is higher here than it is here, the attractive force pulling this to the rod is stronger than the repulsive force trying to push this away. In any case, let's say you rub this glass rod with dog's fur, you have a positive charge. You do the next obvious thing, you rub it with cat's fur and you get a negative charge. It's very early and anyway, this is a negative charge; you notice that when I won't go from positive to negative, let me do this again. That's positive, look at the piece of paper and where the charges are, I change the polarity, I go to negative, the charges move. The particle is still attracted to a region of highest electric field. This is the basic principle of how to hold on to a neutral particle. If you then say in space, absent any material how can you get a high electric field? It turns out you can get it by focusing a laser beam. You cannot get it with a static field, but you can get it with an oscillating field. At the centre of a focused laser beam you have a region of highest electric field, but the trouble is in order for this trap to hold on to atoms, the atoms have to be moving very slowly. Suppose you have an atom moving very quickly. If you surround it two sides with laser beams and you tune the laser beams so that the atom going towards the right, preferentially scatters more light from the right hand beam the left hand beam, because it's tuning itself into the resonance and it's tuning itself out of the resonance due to a Doppler shift. Art Schawlow explained Doppler shift very well, he says, "When I'm walking towards you there's a Doppler shift in higher frequency. And walking away from you, you get a lower Doppler shift." Anyway, that's the Doppler shift. When you preferentially scatter photons from the right hand side you slow up the atom. If the atom happens to be going in the opposite direction, the same thing happens: you scatter more photons from the left-hand beam pushing it to the right. That's the essential thing, no matter which way the atom was going it is slowed down. If the atom happens to be going in a different direction you put in these two beams, it's different. This is a great idea, I got very excited about it, rushed. I told my director, I'd like to drop what I'm doing and do this - another incomplete. He had shut down this work several years before at Bell Laboratories, he said, It turned out that this wonderful idea was stolen from me ten years before I got the idea. Actually Ted Hänsch and Art Schawlow had proposed this idea in 1975, it was a two and a half page/two page paper. Thank you Ted for not doing it. Anyway, I didn't invent the idea but I did invent the name, we called it optical molasses. This is what the experiment looked like, it was a pretty elaborate vacuum chamber, UHV chamber. You see the yellow beams, those are the beams used to make the optical molasses. That green dotted beam is a pulsed laser that evaporated bits of sodium off into the chamber. The way you make these pictures, you just take a white card, the room's dark and you just move it. That's why the pulsed laser looks like a pulsed laser. To you students year, I have to tell you the time scale of this experiment. The time I got the idea I did not have this vacuum chamber. Ten months later I sat down to write the first draft of the optical molasses paper. Whole thing in less than ten months. Okay? I don't know why it took me so long to get a PhD. This is a picture of the optical molasses. When you stick a camera inside the vacuum chamber you see this orange glow of atoms going about 200 Micro-Kelvin. The following year we were able to use an optical tweezers to hold on to the atoms. The year after that we were able to generate a hybrid trap that could trap 10^6, 10^7, 10^8 and even 10^9 a billion atoms. That magneto-optic trap became the workhorse trap of all the people that followed. I should say I also blew it in this experiment. There were early indications that the temperature was much lower than theoretically expected. I sort of fluffed it off, didn't take it seriously. And then a few years later Bill Phillips and his colleagues discovered our mistake, that the atoms were indeed much colder than what theory really told us to expect. Not a little bit colder, maybe five to eight times colder. Again, lots of failure. I moved to Stanford University in the fall of 1987. That's what I looked like when I was thirty nine, I used to be younger. In any case, one of the first early experiments we did was making an atomic fountain. This is where you cool the atoms, you toss them up, they turn around due to gravity. And while they're in free fall going around the gravity, you can actually do very, very good spectroscopy, microwave spectroscopy on these atoms. This so-called atomic fountain led me and a number of other people to develop better and better atomic clocks. Since then there have been other advances in atomic clocks. It's either you hold them with ions. But the biggest advance besides holding, trapping atoms or neutral atoms with laser techniques is the ability to count optical frequencies, something that Ted Hänsch and Jan Hall got a Nobel prize for recently. The progress is remarkable, they're soon to achieve relative fractional frequency uncertainty of eighteen decimal places, 10^-18. If you started one of these clocks when the universe was born - and it didn't get fried - and said, What time is it today?, your uncertainty would be about one half of one second. Why would you want that many decimal places? Isn't this crazy? It's not, we actually want three or four more decimal places because with these exquisite clocks you can do all sorts of things. Indeed you can look at the relative change of the nuclear forces with the electromagnetic forces, and perhaps see changes in fundamental constants in a laboratory time scale rather than a universe time scale. We also introduced what are called atomic fountain interferometers, where in a few years we demonstrated we can measure the acceleration due to gravity with a precision of eleven decimal places. The student who started this Mark Kasevich is pushing this onto greater heights and is improving that by about five orders of magnitude. Overall ten orders of magnitude from the first experiment we did at Stanford on this. Now, when I got to Stanford I thought, we can hold on to atoms. Art Ashton showed that you can hold on to single cell organisms like bacteria. I said, "Can we hold on to individual molecules with light by gluing little plastic spheres onto the ends of DNA." This is the setup where you introduce the laser in the optical microscope, this is what it really looked like. I started working on this in 1989 with a MD PhD student who told me enough biochemistry so I could glue on these polystyrene spheres onto DNA. This is what it looks like. This is a single strand of DNA, looking in an optical microscope. You control a mirror with a joystick and it's like a video game. In fact for three days they played this video game. (Laughter) I'd walk in the lab and say, "I think we really should do some science." (Laughter) Actually the first person after Steve Kron who actually did this was ... He did it as undergraduate student, Steve Quake, who then returned back to my lab his last year of graduate school and stayed three more years as a postdoc before going to Caltech, but we got him to Stanford. These are some of the early people, Doug Smith, Tom Perkins on the polymer work, Xiaowei Zhuang, Taekjip Ha - a host of others on the biology. I don't have time to talk about that, but there are a whole host of very good people. Here's a few more. That's me, also when I was younger. The one on the far left Cheng Chin, he's now a professor of physics at Chicago. Vladan Vuletic at MIT. Hazen Babcock, a brilliant student but shy, married Xiaowei Zhuang. Xiaowei Zhuang was a not so shy person, Hazen was shy. But he is the technical heart of a lot of what she does at Harvard, very gifted scientist. And then Jamie Kerman who's now at Lincoln Laboratory. So these are a smattering of my group. In 2004 I was asked to direct Lawrence Berkeley National Laboratory and did what every self respecting scientist would say, "I'm not interested." They asked me again, I said, "No, I'm very happy at Stanford." Then finally they asked me a third time and the person, my former director at Bell laboratories who was then director of BL said, I came, I visited, they offered me a job; I took it this time because I thought, if I could get really spectacular scientists at Lawrence Berkeley Lab interested in energy it would be worth it. This was a very distinguished laboratory, thirteen scientists who worked at LBNL have earned Nobel prizes, but the thing I'm most proud of is that over thirty young scientists, PhD, post-doctors starting careers as scientists, worked at LBNL, later got Nobel prizes, including me, Saul Perlmutter, George Smoot. We were all graduate students at LBNL when we were at Berkeley, and we're all here at this conference. It's an outstanding national laboratory. I was there for nearly five years and they did get more interested in renewable energy. I'm very happy about that. November 2008, the then president elect, Obama, I got a phone call, said he would like to meet me in Chicago to talk about a job. I said again, "I don't think I'm interested." I was thinking of stepping down from directorship going back into the laboratory. He said, "No, no, no, this is a very important job." I said, "How important?" He said, "Secretary of Energy." Anyway, I flew to Chicago, met him for about an hour just one on one. After that interview or chat I came back and told my wife, "If he's going to offer me the job I will accept it." He did, I did and I spent the next four and a third years in the Department of Energy. I introduced a few things. A new founding programme called Advanced Research Project Agency for Energy, ARPA-E, it's short time, high risk. I wanted to group enough scientists and engineers in a critical mass that can take a problem for maybe the fundamental needs of the problem to engineering technology, we call them Energy Innovation Hubs. We took the photo-voltaic and photo-thermal programmes and tried to re-energize it. Let me give you an example, what were we doing at ARPA-E? Suppose there's an existing technology, the horse and buggy, and as time goes on it gets better and better for the effective cost of this. Then something comes along like a steam powered car but it doesn't really make it. Another car comes along, the Benz Motorwagen, but it doesn't make it. Not because it didn't work, it worked very well, but because it was too expensive, kind of like a Tesla S1. Works very well but it costs a hundred thousand dollars. The car that transformed the technology was the Model T, it was a good car and a lot of Americans could afford it. We call this a disruptive technology. ARPA-E was designed to invest in disruptive technologies, knowing full well that nine out of ten would fail. That was a different funding programme than we had. We weren't looking for safe investments, we were looking for homerun investments. SunShot, we looked at the price of solar for utility scale solar in 2004. The whole system cost 8 dollars, by 2010 it's 3.80 We said, "Where could the price really be?" We set this crazy target by 2020 that the whole system, the modules, the electronics, the land use - everything would be a dollar per watt. That's a dollar per watt of generation based on a certain irradiance of solar energy. They thought we're smoking something - industry - and said, "You are crazy, we may get to 2.50 but not a dollar." We said, "No here's a business plan, here are the things that we can do this." A year and a half later one of the real gurus of solar technology gave a speech and he said - sorry, I'm destroying your podium. (Laughter) Will you invite me back? (Laughter) Anyway, a year and a half later they said, "You know, you're right. We can achieve this. We've redone our business models." So that was very good. In starting ARPA-E and SunShot and a few other places, the idea was we would get very, very good people. As good or better than the scientists or engineers they were funding and work in the department. ARPA-E showed that it really could be in a federal bureaucracy. SunShot, which was an existing programme, if you import three or four people at the top and they're good at leading people into a new vision, you can actually transform an existing bureaucracy. Our motto then was something I say to my students all the time which is, "The greater danger for most of us lies not in setting our aim too high and falling short, but in setting our aim too low, and achieving our mark." That was said by Michelangelo, but it's a good motto for scientists as well. I was tasked by the President to help BP stop this oil leak. We didn't have any jurisdiction but I had made a suggestion to the BP engineers that they eventually, after laughing at it for two days, eventually took the suggestion. At the end of a cabinet meeting shortly after that he goes to me, he says, "Chu, go down there and help them stop that leak." I and a team of five or six scientists spent the better half of that summer down in Houston really getting into nitty-gritty, I don't have time to talk about it but that was something. In hindsight, my most important role at the department of energy was to really identify great people, call them up and say, Once they're there, don't leave them alone, but block and tackle for them because the overwhelming government bureaucracy could drag them down. As one veteran of the Department of Energy said, "How can you help stop the mind numbing, soul sucking BS?" He didn't say BS but that's an acronym. The part I hated worst about politics was the newspapers. They were there waiting, trying to catch you to make a slip, that you would disagree with the President - something like this. Six days after it was made public that I was going to be stepping down, then the Onion ran this story: I read you a part of the byline: awoke Thursday morning to find himself sleeping next to a giant solar panel he had met the previous evening. According to sources, Chu's encounter with this crystalline silicone solar receptor was his most regrettable dalliance since 2009 when an extended fling with a 90 foot wind turbine nearly ended his marriage." We normally don't answer scurrilous reports but I walked in that morning and said, "We have to answer this one." That noon I issued the following press release, "I just want everyone to know my decision not to serve a second term as Energy Secretary has absolutely nothing to do with the allegations made in this week's edition of the Onion. While I'm not going to confirm or deny the changes specifically, I will say that clean, renewable solar power is a growing source of US jobs and is becoming more and more affordable, so it's no surprise that lots of Americans are falling in love with solar." That was fun. After the end of April 2013 I decided to go back to Stanford, I just wanted to be a professor. When I got there I was getting interested in neuroscience. A couple of us got together and said, "There's a number of things we could do." I remember in September 2013, I got an idea. I was talking to a long-term collaborator, Axel Brunger, worked with him for fourteen years on neuro-vesicle fusion. I said, "Axel, this might really be a neat idea, it might work. Let's go talk to Tom about it." That's Tom Sudhof, we got him in his office and I explained the idea and he said, "It might work, it might not work." And I said, "Tom, don't you understand? If this works we might become famous." Two weeks later he gets a Nobel Prize, too late. But he's a very serious scientist and he's now back in the lab. And so we want to study the brain, an operating live tissue of whole brains, and Liqun Luo also. Here's one example of some of the things we want to do, it's a technical thing. In following the footsteps of Neher and Sakmann who invented the patch clamp technology, they really revolutionized neuro-physiology. We wanted to say, "Can we make fluorescent probes that could be exquisitely sensitive, like a patch clamp, in time but where you can look at 1,000 or 10,000 neurons in real time in a live brain, a live interacting brain?" The particle we were investigating, there was a number of particles, this is one. It's a nano diamond grown from molecules of diamond. We're developing this process which is a plasma growth starting with essentially molecules. On the picture on the left is a large diamond, one micron in diameter, where you see the facets. The picture on the right is a smaller diamond, maybe seven nanometers in diameter where those lines are and you can see the facets. The one on the right is a perfect diamond. It's something you would want to put on your finger and wear proudly, if you can see it. We've learned how to dope them with independent control using silane gas, again in a CVD process. That particular diamond emits10,000,000 counts a second. We think we're now learning how to 'mass-produce' these in large quantities: We're also trying to figure out if we can make these photosensitive. So that if we can embed another nano-particle in a neuron, in the membrane of the neuron, and there's an actionpotential spike, the fluorescence of this will change. But since they are photostable and give out so many counts we can see individual actionpotentials and they should last ... We want them to last as least as long as the graduate student. We're learning how to functionalize them, this is non trivial. This is some work now only one week old, where we can put SiO2 coating on these nano-diamonds, only 1 nm thick. That's the first step in functionalization and then we're going to have to make them hydrophilic and go and and go forth. Stay tuned, it's stuff. I'm running out of time, I'm not going to really have time to talk about a different class of particles, rare earths. You can dope in a single particle different ratios of colours, so that in one particle can be this colour, in another particle could be these two colours. And in another particles could be this colour and another particle could be this colour, and another particle can be this colour. If you consider all the colour combination that we've already synthesized, in principle you can have 39 colours. Then what you can do is you can take the image, you can split one off to do 3D-localization and the other one you put in a prism to disperse the colours, so the array detector begins to be your spectrometer. And by looking at the ratio colours you can in one frame grab many different colours. Then you can go very rapidly in doing this. Of course, this means that these have to specially dispersed. Let me end by saying why I went for ten years into something I cared deeply about. If you look at this image, this is taken by an astronaut December 24, Christmas eve 1968. He said, "We came all this way to explore the moon and the most important thing is we discovered the Earth." Since 1968 we've discovered that the climate's changing, in large part due to humans. If you look at this picture, you don't have to be a rocket scientist to realize the moon is not a good place to live. From this vantage point the Earth looks pretty good and there's nowhere else to go. Thank you.