SAUL: I'd like to welcome you to the 2017 Bethe Public Lecture. This lecture series honors Hans Albrecht Bethe, who was a true giant of 20th-century physics. Hans was born in 1906 in Alsace-Lorraine He moved to Germany and took his PhD at the tender age of 22. With the coming of the Nazis in Germany in 1933, the Bethe family first fled to Great Britain, and then in 1935, came to Cornell University where Hans remained until the end of his life.
Over the course of his long life, Hans used quantum mechanics to, really, in many ways, completely change the way we understand our universe. He was unusual, even amongst great physicists, in the enormous breadth of the things that he touched. Nuclear physicists, astrophysicists, solid-state physicists all lay claim to the fact that, in many ways, he was the founder of large parts of their field. For example, within four years of coming to Cornell University, Hans explained the longstanding puzzle of how the sun can shine for billions of years. And in fact, his explanation of nuclear fusion going on in the Sun is what won him the Nobel Prize in 1967.
He was very much one of a kind. He published at least one of the most important papers in many fields in each decade of his long life. And he lived quite a long time. He passed away at age 98 in 2007. And the paper that he published in 2005, two years prior to that, is still considered a very important paper.
Hans's impact on humanity was really profound. And I really would say that it goes well beyond just the sciences. During World War II, he led the theoretical physics division at Los Alamos, the Manhattan Project, out of fear that the Nazis would first get a nuclear weapon. But when Germany fell, he turned with a vengeance to arms control.
With incredible integrity, courage, and tenacity, for example, he fought against the insanities of the McCarthy era in this country. He helped persuade President Kennedy to sign the 1963 limited Test Ban Treaty, which forbade testing of nuclear weapons in the air, water, and in outer space. He helped persuade President Nixon, in 1972, to sign the treaty that halted a then-growing anti-ballistic missile arms race. He was a formidable opponent of Ronald Reagan's Star Wars program in the 1980s by arguing very persuasively about the unfeasibility of what was being proposed.
So in many ways, Hans Bethe was an astonishing man. Perhaps more than any other person, he was responsible for creating the culture of collegial understanding, and yet rigorous scientific understanding, and social interaction that's a hallmark, I think, of the Cornell physics department.
My colleagues and I are eternally grateful for the gift of Bethe as a member of our department. So with that, I'd like to turn to the introduction of the 2017 Bethe Lecture. And it's my great pleasure to introduce Professor Margaret Murnane.
Hans would have been thrilled by the way Margaret's recent work on tabletop X-ray lasers, which I think you'll hear something about today, opens completely new domains of science. Margaret received her bachelor's and master's degree from the University College Cork in Ireland-- she's Irish-- and then in 1989, her PhD in physics from Berkeley. Today, Margaret comes to us from the University of Colorado, Boulder, where she is distinguished professor in both the physics department and electrical computer and energy engineering. She's director of NSF Strobe Science and Technology Center, which is a center devoted to functional nano imaging. She's also a research fellow at JILA, which is the world-famous joint Institute at Boulder between the University of Colorado and the National Institute of Standards and Technology.
Margaret runs a joint-research group at a small company with her spouse, Professor Henry Kapteyn. In the interest of time, I won't go through Margaret's many honors. But you got a sense that there are many. In fact, she is an extremely accomplished and distinguished scientist, as I think you'll appreciate very shortly.
Margaret's work pushes the physical limits. Yesterday, she gave a talk on laser experiments on the 200 attosecond timescale. That's 2 times 10 to the minus 16 seconds. To give you a sense of how short a time interval that is, 200 attoseconds as to 1 second, roughly is one second is to 160 million years. It's a really short timescale. So with that, please join me in welcoming Professor Margaret Murnane.
MARGARET MURNANE: Thank you very much, Saul. It's a great honor for me to present the Bethe lecture series here at Cornell and to talk to you this evening about the science that I do with the big research team at Boulder. But I also need to thank the Cornell community for making me feel very welcome over these last three days. And I really want to thank you for this invitation.
So this evening, what I'd like to tell you about is a little bit about myself and how I came to be a scientist, and then a very unlikely answer to the challenge of trying to make a tabletop X-ray laser. It's a very unlikely result. But it turned out to be quantum mechanics that saved the day and actually made it possible to extend laser light to short wavelengths for the first time.
As Saul said, I grew up in Ireland. And as I was walking around today in the beautiful sunshine, I realized that Cornell is very much like the area I grew up in near a river, really some gorgeous vistas. And I think it was growing up in that beautiful environment that really attracted me to light.
I was the first person in my family to be able to attend college. So I didn't have anybody to inspire me to do science. But I was very fortunate that my father was a grade-school teacher. So he would bring beautiful books home. Of course, this was way prior to the internet.
And one of those books, when I was about nine years old-- are there any kids in the audience? Or-- those of us who pretend we're still kids, OK. And in reading one of those books, I was just fascinated by the story of Archimedes. I was about eight or nine years old.
And two things fascinated me about this story. First of all, I could kind of grasp the concept-- probably not all that well, but I thought I grasped it-- this idea that-- you're trying to tell if the crown was pure gold or had other metals inserted. But you could tell which was the golden crown by just the amount of water that the crown displaced.
So the concept was attractive. But the other thing that got me really excited was the fact that Archimedes was excited by an idea. Growing up in Ireland, we were very excited by poetry-- Yeats, literature by Joyce, music.
But at the time, this idea that an idea might excite you was something I just fell in love with. I had never met a scientist. I decided I was going to be a scientist when I grew up.
And, of course, very fortunately, as I grew up, Ireland started changing from agricultural to more slots available for university. So I was very fortunate to be able to attend University College Cork. And I met my first scientist when I was 18 years old.
So I arrived into the physics department, and I informed my freshman lecturer that I was going to be a physics professor. And [INAUDIBLE] had a great sense of humor. And he said, well, maybe you shouldn't put all your eggs in one basket. I didn't even realize what he was telling me. So I proceeded on.
And I was very fortunate. I met my first female physicist when I was 21. And I learned that you could actually get paid to do science as a student. This was a very new concept. And I was very fortunate to do a master's degree that really introduced me to the beauty of light-- something I had been fascinated in since I was a kid.
And then just to illustrate how different Ireland is now from when I was an undergrad, when I attended university in Cork, there were 5,000 undergrads in the university. And now in that same school, there's 25,000 and a PhD program that is just world-class-- so, just tremendous changes.
But at the time, actually, there was no way to do a PhD in Ireland if one wants to become a scientist. And so I left Ireland. My first trip abroad, I was admitted at Berkeley and started graduate school there. And because I hadn't traveled in other areas of the US, for the next seven years, I thought every place in the US was like Berkeley. So it was a big surprise to me to find out this was not the case.
So what did I discover? I discovered I loved lasers. As a kid growing up, I had terrible trouble sewing or painting. But who knew it-- I could align lasers really well.
And the chances of actually figuring that out-- think about it. And it was just something about that eye-hand coordination that I had. And so I loved lasers.
I discovered I hated plumbing. So that that definitely was a career path I was not going to go. And the reason I discovered that is that the kind of lasers that we were working on were really-- were die lasers, you had to build circulating systems and flow the die at a tremendous rate. And so I knew I did not want to do more plumbing.
And I was very fortunate that the second student hired in my research group, Henry Kapteyn-- we got married during graduate school, and we've worked together ever since. So I don't think there's anything better than sharing a eureka moment as 1:00 AM-- sharing it with somebody who knows you. And I think all of the faculty and students and Cornell can tell you that, as scientists, we may not look social. But we are super social. And we love to talk about what we're doing with each other.
So as we were working together, Henry told me-- Henry second name is Kapteyn-- my mom and dad are Dutch, so "Cap-tine" and he said, did you know there is a Kapteyn's star? And I thought he was joking me. But no. In fact, Henry's great uncle is a very famous astronomer who actually has the distinction, among many things-- there's a Kapteyn series, there is a Kapteyn's star in the southern universe-- but also, actually, wrote a paper where dark matter was mentioned for the first time.
At the conclusion of that paper was that dark matter did not contribute. So the paper wasn't right. But it did mention it. And there's actually a Kapteyn Institute in [INAUDIBLE] in Holland.
So we were Roger Falcone's first two students, moved as faculty to Washington State University. And the wonderful thing about that university is that the physics department was very small. And people knew you couldn't do hard science alone. So they allowed us to work together, which is very unusual for junior professors. And that's where we did the laser work I'm going to tell you about.
And then we got a great opportunity to move to the University of Michigan. And I always felt we were in the wrong place at the wrong time, because in Washington State University, we were in the physics department doing very fundamental light science. Then in Michigan, we were in the physics department doing sort of laser science-- developing technology. And then in Michigan, we were in an engineering department doing very fundamental science.
And now in Colorado, I'm not sure what we're doing. We're doing laser science, materials science, nanoscience, chemical science. And it's a real, I would say, richness of science in the 21st century, that the divisions are being blurred, that not only is it fun to talk about materials science, or chemical science, and nanosciences, but we can do more by working together. And so there has never been a better time to be a scientist, in my opinion.
And so now we are in Colorado. We love it. The precipitation is mainly in the form of snow.
But I think the best part about my work and my job is not only the fact that I love science and discovery, but also, it's the people that we have, really, the pleasure and honor to work with-- a group of students from all over the world. They're usually the happiest when they are graduating. This is several graduates.
Mark Siemens is now professor at the University of Denver. But Mark joked that he thought his daughter was the only one who really read his thesis very thoroughly. You might see, this is a laser beam bouncing off a particular grating. But they look very happy there. They can be tough, because we do live out west.
This is of some of the group. This is a relatively old photo. But this was taken in the four corners area. There is a steam train you can get from Durango to Silverton. And it's a really beautiful area, particularly this time of year, just like upstate New York-- really gorgeous fall colors.
OK so that's just a little bit of background on how I came to be a scientist. And so what do we do now? Essentially, trying to understand and master light. Of course, light is how we understand our natural world. We look around. We see and make sense of the world through light.
And you can make an argument that our modern understanding of nature began with Maxwell's equations-- understanding how light propagates. And it's sort of shocking to realize that at that time, time and time again, we think we've come to a full understanding of nature. Thank goodness we've always been wrong-- and so Maxwell said, after writing down Maxwell's equations, in a few years, all great physical constants will have been approximately estimated. And the only occupation which will be left to men of science will be to carry these measurements to another place of decimals. And thank goodness he was wrong or else we wouldn't be here.
So it was light that really showed us the quantum nature of matter. It was the photoelectric effect that Einstein explained that you shine light on a material, and the electron energy that is emitted has a threshold, and it depends on the color of the light. So that showed us that light comes in bundles called photons. And that led to our understanding that quantum mechanics underlies all of nature. So we can say modern physics is overwhelmingly based on the discovery of the quantum nature of light.
And I'll try and tell you a story about now, quantum dynamics in atoms is giving us an ability to control light in the X-ray region of the spectrum that we never thought we would have. It's mind-boggling what we thought 20 years ago compared to what we can do now in terms of controlling light. So why is this important?
Well, it's important because in a coherent light, or directed laser-light beams, are tremendously important for science, technology, and our economy. And when the laser was first demonstrated back in 1960, it was initially called a solution looking for a problem, because prior to that, we had light bulbs, or incoherent light. And so all of the applications-- none of them were optimized for taking the directed beam of light and harnessing that capability.
But over the decades since, lasers are used to power the internet, for welding, surgery, holography, data storage-- just a tremendous amount of applications that really drive science and the economy. And the power of lasers, their capabilities, has also improved tremendously, advanced tremendously. But in the 57 years that laser light has been around, there's one area where they have not improved very much. And that is in terms of their wavelength.
The first laser that Namen demonstrated was in the red-- the ruby laser. And the shortest-wavelength laser that's now in widespread use is the excimer laser 193 nanometers. The ruby laser is about 700 nanometers. And so that advance in wavelength is actually less than a factor of 4 in 57 years.
And in almost every other area of science-- in data storage, electronics, all other areas-- the capabilities have advanced by many, many orders of magnitude-- 6 or 12 orders of magnitude-- but not in our ability to make laser light at short wavelengths. And the quantum light sources I'm going to tell you about are the only current way that we know to make X-ray laser beams on a tabletop.
Now, we have X-ray sources better in widespread use. You meet them in your doctor's, or dentist office, or in security. And of course, X-ray diffraction was used to uncover the double-helix structure of DNA. So x-ray sources based on the X-ray tube have widespread impact.
I don't know if you looked at your dentist or your radiologist-- try to read an X-ray. And that X-ray is not a whole lot crisper than that first X-ray that Rontgen took up his wife's hand. And the reason that that is true is essentially, the newer X-ray tube technology is brighter than that initial X-ray tube that Rontgen used. But essentially, it is the same basic technology. It's essentially an X-ray light bulb.
You might say, well, why not use an x-ray laser, and that would let us take an image, whether it be of trying to diagnose cancer or for whatever application. Why not build an x-ray laser?
Well, it turns out that it's energetically prohibitive. So if you want to build a laser, you need to put atoms in an excited state and keep them there for a relatively long time so that you can have what's called stimulated emission. I'll show a movie on that in a moment. And if you're putting atoms in a very highly excited state, they'll tend to de-excite very quickly. So you need an enormous amount of power to put the atom in a very highly excited state. And in fact, that power scales as the wavelength to the inverse fifth power.
And so what that means is that I can hold a laser pointer in my hand, because I only need a pair of triple-a batteries to power it. But if I want to make a one nanometer wavelength, or 10 Angstrom wavelength x-ray laser, that would take a terawatt of power. And that is on order of the electrical generating capacity of the United States. So there is no way to have an x-ray laser just there for, let's say, a medical diagnostic or something. We just can't do it. It's too power prohibitive. And if you want to make a one Angstrom laser, you need 1,000 times more power, a petawatt.
And that's why, as Saul mentioned, during the 1980s during the Reagan era, there was an initiative called Star Wars where one of the ideas was to put an x-ray laser in space to shoot down missiles, and powered by a nuclear weapon. And of course, that is not scalable. Or
And so it's for that reason that the first coherent or laser like beams in the x-ray region were built on a completely different technology based on the beautiful x-ray accelerator work done here in Cornell where one accelerated an electron beam over a few kilometers to make what's called a free electron laser. Wriggling high energy electrons can give a beautiful laser like x-ray beam.
Fantastic for basic science, but for widespread applications, it's also a good thing to see what aspect of these very beautiful and powerful machines we can have on a table top. We won't be able to have the same power, but some of the capabilities we can have in a tabletop scale apparatus. So we can apply them in medicine or nano manufacturing or other materials growth and optimization process.
So if we can't use x-ray lasers and we want to find a smaller technology than the electron accelerators, what can we do? So we can make an analogy with sound. And if you think about a stringed instrument, guitar, violin, and you pluck a string gently, then you're going to hear the fundamental tone.
If you pluck it harder, you'll hear maybe the higher harmonics. And if you pluck it very hard, you might be able to hear quite high order harmonics. Particularly if you're a very good musician, you can do that. There's a limit to doing this, because the higher order harmonics tend to be dimmer and dimmer, weaker and weaker.
So can we do this with light? Yes, we can. And that experiment was first demonstrated only one year after the laser was demonstrated. Peter Franken took the ruby laser and focused it into a quartz crystal. Even politicians talk about focus like a laser beam, and they do this for a reason.
Because at the focus of a laser beam, there's so much power that we can drive electrons in a material non-linearly. And that allows us not only to transmit the fundamental laser color, but also make the second harmonic. In this first experiment, they didn't know how to make this light very bright.
So there was a little speck corresponding to blue light on the photographic plate when Franken submitted this work for publication. And in fact, the typesetter thought it was a speck of dust and whited it out.
In this first paper, there's no experimental evidence. There's nothing under that little arrow. But they never published and erratum. Because if you had a ruby laser, you could focus it on anything, pretty much, and with your own eye, you could see the blue light. So everybody believed you could actually change the color of laser light from one color to another. So everybody believed in this what's called harmonic generation.
Just an analogy with sound. So we hear harmonics in sound. We can also see them in light. And what one is doing is taking two red photons to make a blue photon. But in exactly the same analogy with sound, the higher harmonics are weaker and weaker. The same was true with light. The higher harmonics were weaker and weaker.
It was barely practical to use the third harmonic. And that just got you from the visible to the ultraviolet. Didn't get you anywhere near the region that you were really trying to get to, the vacuum ultraviolet or extreme ultraviolet, or x-ray region of the spectrum.
30 years later, finally got a better handle on how to make very intense lasers. This is just a silly picture from the '50s. So to explain how we made intense lasers, let me first just remind you of how atoms emit light. You can excite an atom with a photon or an electron or chemical energy, excite an atom into an excited state.
You can think about atoms being lazy. When you excite them, they always want to go back to the lowest energy state. But after you excite them, they'll de-excite and emit a photon in a random direction at a random time. Kind of like a light bulb.
To make a beam, what you have to do is force the atoms to de-excite at the same time. And there's nothing better than a laser to be able to do that. And in fact, here is our simulation. So let's slow it down. Reset. And so we've got excited atoms in a laser cavity with two mirrors and one that's partially transmitting.
So these are the photons traveling back and forth, they interact with the excited atoms and force the atom to de-excite and emit that energy as a wave that is perfectly aligned with the crest perfectly aligned to the waves that exist in the cavity.
And this is what people mean by coherence or a laser like beam. And so then you get a beautiful-- this beam can be continuous or come in bursts. But the significance is that all of the waves are in the same direction with their crests aligned. And that's what gives the directed nature of laser light.
Now, one common misconception is that laser light has to be a single pure color. That's true of many lasers, but not of every laser. So often, we think it's either a single color of red or green or whatever. And that would be a single sine wave. We have lasers now that can emit many different colors.
And if those waves are synchronized and you add these waves of many different colors, what you will get is constructive interference in one place and destructive interference everywhere else. So you will get a sequence of pulses out of the laser. And I can show you a little animation that illustrates this, perhaps. So that's shown here in this simulation where what you see now is the emission of a continuous wave laser emitting a very pure color.
Now, let's look instead at what the output might look like if you have a laser that can emit many different colors, but that are phased with respect to each other. Then what you get is a sequence of pulses. Where all of the waves interfere constructively, you get a pulse. Where they interfere destructively, you get nothing.
And so that gives a sequence of pulses. And the more waves you add of different colors, the shorter the pulses that come out. And essentially, these lasers, these short pulse lasers, gave us the means to generate these very short bursts of light in the x-ray region.
And so back in 1986, Peter Moulton was working at Lincoln Lab adjacent to MIT. And he made a beautiful laser material called titanium doped sapphire. It's very similar to the ruby you know from jewelry. Ruby is chromium doped in sapphire. Titanium sapphire is titanium doped in sapphire. So it's a little bit pinker than ruby. And Peter developed this material for space based lasers. It could store a lot of energy. So in a very small crystal, one could get quite a lot of laser energy.
But the other distinguishing factor about this laser is, even all these years later, it is still the laser that can lase over the most colors of any laser material that we know, which is kind of remarkable that that record has held up for so long. So it can make laser light all the way from the red into the near infrared. And in a very simple way, all of those colors can be synchronized perfectly. So in a very simple set up, and so this was the work Harry and I did as assistant professors, we could figure out how to make a 10 femtosecond light pulse.
And as Saul said, these time scales are sort of so fast, they're hard for us to imagine. So one way to put them in perspective is the geometric mean between 1 minute is the geometric mean between a 10 femtosecond light pulse and the age of the universe. At the time, we thought they were so ridiculously short pulse, that there was no reason to want anything shorter.
And if you actually looked at the time scales in nature, if you use a shutter to take a series of images as Muybridge did in this series of images of the galloping horse, you're capturing a millisecond event. If you look at the beautiful work of Harold Edgerton that captured a bullet penetrating an apple, that's a microsecond event. Then to go to shorter time scales, you have to go to smaller things.
And so for example, molecular rotations happen on nanosecond to picosecond. A vibration in a molecule can happen from hundreds of femtoseconds down to 10 femtoseconds. And once you get to a femtosecond, only electrons can move any significant distance. And at the time, this was back in 1990, so almost 30 years ago now, and because it's very hard for us to project forward and think what would we need 30 years from now?
And so a colleague, Wayne Knox, who's now a professor in University of Rochester, at the time he was working in Bell Labs, he wrote this article for the April 1 issue of the Optics and Photonics News. His dad was [INAUDIBLE] Knox and he had a colleague, [INAUDIBLE]. And many of you may know [INAUDIBLE]. So the hint is Knox Knox, who's there.
Because at the time, this was before people realized that in a chemical reaction, there is actually a transition state that happens for a very short time, that there was something there that you could use a strobe light to capture. People didn't know that back then. So people thought that these 10 femtosecond light pulses were so fast that they were actually useless. Anyway, so Wayne wrote this silly article saying, we're investigating possible violations of thermodynamics. Somebody's pulses must be getting longer.
So why am I talking about this? It's because these short light pulses, as well as uncovering the transition states in chemical reactions, are also, you could think about it as the best guitar pick for making light. Think about it. If you try to make higher and higher harmonics of sound and you pluck that string harder and harder, eventually the string would break.
But if you have a 10 femtosecond light pulse and you hit an atom with that 10 femtosecond light pulse, the atom does break. It ionizes. But it's fine, because there's a whole 10 femtoseconds to play with the electron, because it takes a little longer for the electron to leave the atom. So that's why these short light pulses turned out to be the very best, let's say not so much hammer, but light to use for high order harmonics.
And they were discovered by accident. People had, over the 30 years or so since the laser was invented, scientists had sort of given up on being able to go much further than the ninth harmonic. They figured they were doing well getting there. But Charlie Rhodes and his group in Chicago focused a short light pulse into a gas and were very surprised to see that they saw harmonics up to the 17th order.
That was a big surprise. It just was not expected. And very soon after that, people saw, this was work from Steve Harris's group at Stanford, saw harmonics up to the 100th order. And that was 10 times further than anybody had even been thinking about. It took about another six years to figure out theoretically what was going on.
But we now know what was going on. The femtosecond light field is so strong that it literally just bends the Coulomb field that's binding the electron to the atom and just causes the atom to partially ionize. It plucks part of the electron from the atom. And that can happen because of the weirdness of quantum mechanics. So you can pull part of the electron away. And then when the light field reverses, the light field can push the electron back to the ion.
And then any kinetic energy that the electron has can be converted into an x-ray. And this happens on every half cycle of the laser. And what you get is a beautiful harmonics of light. And these harmonics now, more than 30 years later, now extend up to the 5,000th order. So it's by far the highest order non-linear process that we know of.
And just to show you what the weirdness of quantum mechanics looks like, so that you can imagine this movie as what's happening to the electron in hydrogen. It's getting driven back and forth by the laser field. And these quantum modulations give rise to these high order harmonics being generated.
The other way you can think about it is that you can think about the atom in a strong laser field as being equivalent to the quantum version of the Rontgen x-ray tube. So how an x-ray tube works is that we boil electrons off of filament, accelerate them in a strong electric field, and then when they hit a target, you get what's known as [INAUDIBLE] emission. That is essentially an x-ray light bulb.
Now, if you take each one of these steps and say, how could I make that quantum coherent? You would have to start with the electron in the ground state, in some coherent state. And that's in the ground state of an atom. You'd have to accelerated it in a coherent field, and that's the laser field that's driving it.
And then it would have to re-encounter a coherent state. And that's because it encounters the part of itself that didn't ionize. And so every step in this process is quantum coherent. And that's why it allows you to take a laser in the visible region of the spectrum and make a coherent beam in the x-ray region of the spectrum.
Took our community a long time to figure out how to do that. We knew the quantum weirdness of this process back in the early '90s, but it took another really 20 years to figure out how to make the beam bright enough to be useful. Part of that delay was that this work had to bring together two communities, people who understood laser light and people who understood atomic physics.
And they went to different conferences, for the most part. So it took a long time for people to talk to each other and not talk past each other. So it's one of those cases, oftentimes, it's good for things to be mixed up, because then we can end the barriers to be broken down so we can make progress faster.
So why did it take so long? So there's a nice analogy with laser light. I showed you the laser cavity and this need, if you want to make a beam, that you have to add waves so that the crests are all aligned, so that they interfere constructively. As the laser light encounters atoms, it generates an x-ray wave.
But if the laser wave and the x-ray wave, if they travel at the same speed in the medium, then sure enough, the x-ray that's generated from the second atom will interfere constructively when we start to build up a nice beam. But it's very hard to do that if you have many atoms in the medium.
Our collaborator, Carlos Hernandez Garcia, made us this cartoon to try to illustrate how hard it is to do that. So what we have here is a quantum conductor. So the laser is acting as this quantum conductor with the baton. And what one has to try to do is force the same number of atoms as there are people on the planet to sing in tune at the same time. And the laser wavelength is 1,000 times longer than the x-ray wavelength.
And so we don't have any electronics that will allow us to do that. And so the only way we can do it is to try to understand the physics so that we can add the x-ray fields with sub Angstrom spatial resolution and then a sub attosecond. An attosecond is 10 to the minus 18 seconds temper resolution.
And over a period of 15 years, our community figured out how to do that. Now we even have a recipe that we can write down and show that what we do is we can use these are different colored driving lasers. And if we use them, the lines are the theory prediction, and the dots are where we've verified these predictions experimentally.
So if we use this color driving laser and this is visible near infrared, mid infrared then we can make x-ray beams that banned from the vacuum ultraviolet through the extreme ultraviolet and the soft x-ray region of the spectrum. The ultraviolet are the lasers that are currently used to print your chip, to print the chips that are in your cell phones and your devices. The extreme ultraviolet is the light that industry is going to try to use to print your next generation circuits as the sizes reduce further and further.
And so practically, it looks quite simple. This is the converter that converts the laser light to an X-ray beam. I'm not showing you the femtosecond laser. That would fit on the size of this podium or this table here. And so we can change the color of the driving laser and then make x-ray light in the extreme ultraviolet into the soft X-ray region of the spectrum.
And the power is microwatts to miliwatts. About as powerful as this laser pointer. But it turns out that that's more than enough to do many applications and imaging and spectroscopy. And in fact, we can make this very broad what we call a coherent super continuum.
In other words, it is coherent light where the waves are synchronized to each other all the way from the visible to the soft x-ray region of the spectrum. And this is the only way to do that. There is no laser that has a bandwidth that broad. So it's an unusual and unexpected ability to make coherent light.
To try to put that in perspective, I showed you this graph before, but we stopped at a femtosecond. Now we have an ability to make or 10 attoseconds or shorter X-ray bursts. If we go to a femtosecond is 10 to the minus 15 seconds. An attosecond is 10 to the minus 18 seconds. A zeptosecond is 10 to the minus 21 seconds.
And in fact, this ability to drive atoms with femtosecond lasers, we have a theory paper that shows that in theory, we should be able to make what's called zeptosecond wave forms. We can sculpt light with features below an attosecond duration. I love the word zepto. I have a cat called Zepto. That's him. He's pretty big though.
And we can play all kinds of tricks driving that electron with linear polarization, two colors, circular polarization. Because we can drive the electron in lots of different we call them trajectories, that gives us the ability to make light with lots of different features. One of those features is now we're able to make these high harmonics with arbitrary polarization with linear or circular, which is very nice for looking, for example, at magnetic materials or chiral molecules with very strange looking waveforms.
I showed you that sine wave that you might think was the normal output of a laser. Pure color, pure sine wave. This thing looks crazy in comparison, but it is a real wave that we've made in the lab and fully characterized it. You can barely read here, but this is one femtosecond, 0 femtosecond.
So we have this light field that's changing in direction and color and amplitude on a sub femtosecond time scale. So clearly, the type of thing we want to look at with these light fields are what electrons do in materials. How do they talk to each other? How do they move in a superconductor? Questions such as that, which are also questions that many people here in Cornell are asking.
So I would say that it is a very unexpected outcome. I started out doing my PhD work where we were very happy if we had a one picosecond burst of incoherent x-rays. That was my PhD thesis work. And we were so excited. Little did we know that we'd be able to make pulses that are six or seven orders of magnitude faster. It wouldn't even have occurred to us that we could do that or what we might do with it.
And now that we know how to control a radiating electron wave function, that was that quantum movie, we have arguably, now not everybody would agree, but we have arguably better control of light in the x-ray region than we have in the visible region of the spectrum. That is something we would not have anticipated. And it's because we don't have to use optics. What we're using is visible light to control quantum mechanics. And through that, control what x-ray light is emitted.
So we can control the spectrum, the polarization, and we can control whether the x-ray light emerges as a single burst or multiple bursts. So it's as if we're using light as a spectrometer or as a polarizer. We're using the visible light as a polarizer or a spectrometer or a filter. And actually, that capability is very useful in the x-ray region, because optics are notoriously difficult to make. You can buy a beautiful camera lens. It's a consumer item. It's polished beautifully.
That same lens in the x-ray region doesn't exist. They're not bent by materials. And you would have to polish it to the same ratio of the wave length to work well. We don't know how to make them. They're very lossy. They're very imperfect. And so being able to control light in the x-ray region using light in the visible region is very useful.
And so now as a compliment to the beautiful sources here at Cornell, Berkeley, Stanford, and elsewhere, we don't have all the capabilities that one can have at the large facilities sources. But we have some of those capabilities. We can reach some of the photon energy range. We can make very nice images.
And we can have them in the lab so that a grad student can make their own discovery in science, not necessarily what their advisor told them to do, which is what really is the hallmark of when one really believes one is a scientist. If you make a discovery that nobody else knows or told you about, and you're the one who made the discovery.
And what we're finding is because we have a new source of light, that anything we pointed at, we learn something new about it. Whether it's a magnetic material or whether it's how do we do thermo management in tiny nanostructures, or what happens to materials if we dope them, or can we make a perfect microscope in the x-ray region?
We now know how to do that because we don't need optics. If you have coherence, it turns out you can make a microscope that has no lenses. So there's no imperfections. And for the first time, we can actually build a perfect x-ray microscope. And then also get insight on how electrons interact with each other in materials.
And so with that, I know I'd like to move to my last slide. And let me end by just giving you a little perspective on how long it takes to do anything in science. So this is just a comparison between the first observation of magnetic resonance in 1937 and the first commercial MRI machine.
It can take 50 years to go from a new observation about nature to when we know how to use that for a new technology. And in this comparison, we're not trying to say at all that these high harmonic sources would have the same impact.
But just to show you another example of where Charlie Rhodes saw these 17th harmonics. It took another six years to even understand how they were being emitted, another 20 years to figure out how to make a beam. And it's only now in the last five years we're actually figuring out that we can do things that you can't do with other approaches. And you know it is definitely leading edge technology, as the graduate students in the audience know what I'm talking about. So this even 30 years after magnetic resonance was observed.
And this was the first commercial NMR spectrometer by Varian. And that first commercial instrument was not a well developed iPhone whatever. It was a painful thing to use. And of course, all the grad students in the audience will empathize with this, because all their projects are like this. Because it wouldn't be research if it was already developed.
Keep cool and say nice things to the machine. Nothing else seems to work. And I joke with my grad students that your PhD thesis will not work until you are about ready to take the whole apparatus and throw it out the window. And then it works just when you're about to do that.
So in conclusion, I would say that sometimes you hear it discussed in science, but many of these problems span multiple decades. You chip away at them and it takes just a long term sustained effort. And it does take a team. Fabulous students, collaborators. And it's really the only way our community was able to make progress. And I want to thank you for the honor of being able to give these lectures and to share the science story with you. Thank you.
AUDIENCE: [INAUDIBLE] The opinion seems to have [INAUDIBLE] in a few years. But he goes on to say, but we have no right to think [INAUDIBLE]
MARGARET MURNANE: Good, good, thank you. And I stand corrected and thank you. But of course, this happens every-- you know, it's a cycle. That's right, yeah. And thank goodness that we still don't know everything. Isn't that right?
AUDIENCE: So is there any reason why at some point [INAUDIBLE].
MARGARET MURNANE: Correct, yes. So certainly we expect to be able to go further. And the reason right now we don't know where the limit is, that if you really want to be able to-- you saw that quantum blob. If you drive the electron in the atom very hard, that quantum blob expands.
And so to actually theoretically predict where the limit was, you would have to do what's called a three dimensional time dependent Schrodinger equation to model that expansion. And then you would also have to couple that with Maxwell's equations. And there isn't a computer powerful enough to do that. So we don't know if the limit is going to be 10 KEV or 50 KEV. We just don't know. And the theorists don't either.
So what we're trying to do now is we make the laser a bit better. I mean, our whole community. So they make the lasers a little better. Then we test, then we work with the theorists and they try to do an approximation so that we can push the theory a bit further. So it's like a little ladder. We're climbing it up with the theorists to try and see where the photon energy limit would be.
But it's already actually quite useful for probing materials and also for making microscopes. Their x-ray microscopes are notoriously not imperfect. And so now there's the chance to do imaging for real world technology applications using these light sources. They're too weak to print chips, but they're fine to look at processing.
AUDIENCE: I'm not fully aware of the technology, but it seems like you have some kind of [INAUDIBLE] to get the [INAUDIBLE]. Is there a limitation to do with that technological step that limits your coherence?
MARGARET MURNANE: So you're right. We have to use gas plumbing, because we need to generate the light in a high density multi atmospheric gas. We contain that in a very small waveguide. And that helps keep the laser beam propagating in a very controlled way. And then we can adjust the pressure and parameters so that the x-ray waves interfere constructively.
We were actually surprised it worked so well. There was just many aspects of the physics that came together. None of us thought that we could do the calculation, a simple calculation. We had to do an approximate theory. And it was just one of those times every so often that it actually worked as predicted. And of course, at some point there may be a deviation and there may be decoherence, but we haven't seen that yet.
AUDIENCE: I'm just wondering, what do people think is going on at the zeptosecond time scale?
MARGARET MURNANE: So if you think about material or such, there isn't. But as we go to the higher photon energies, then those time scales and energy scales would correspond to nuclear processes. But I mean, we're far away from being able to reach that energy scale yet.
MARGARET MURNANE: No, your comment was very good.
AUDIENCE: I'm just curious, are there more prefixes that we have to learn?
MARGARET MURNANE: Oh yeah, there's a yocto. Yeah, there's a yocto.
AUDIENCE: Is there some fundamental, theoretical limit?
MARGARET MURNANE: There's a yocto. Well, the guys who are building the big lasers, Frank does this big. So in another direction, femtoseconds are being used to pack so much power into a small spot that you could do per production from vacuum and such. And so for that, they have to go up to exawatts and [INAUDIBLE] and such. And so I think there'll be a few more named orders of magnitude before we hit a physical limit that you can't either go up or down. Up in power or down in pulse duration.
AUDIENCE: Underlying fundamental physics of the limit?
MARGARET MURNANE: Oh, there a limit in actual power density that you can extract out, put on optics or materials or such. So that would be in power. In pulse duration, I think you have to think about it differently. Think about a waveform generator, an electrical waveform generator in an undergrad physics lab. So what this allows you to do is make synchronized light that spans, say, from the terahertz to the X-ray.
So you can think about manipulating matter. And scientists are already doing this where they can control the lattice vibration and see how the electrons behave in response to a specific vibrational mode. And we could think about that also in terms of just making that control more-- one can control more parameters. So it really is this controlling light in a very precise way over a huge range of the electromagnetic spectrum.
AUDIENCE: Something in terms of H bar and C or G.
MARGARET MURNANE: In terms of the bandwidth, I'm sure there is.
SPEAKER: May I have the last comment?
MARGARET MURNANE: Of course.
SPEAKER: At dinner you pointed out that the high harmonic generation discovery in some sense was serendipity. It was a student doing the experiment that they should not have done because their advisor knew it would not work. Time and time again, we see that's how science moves forward. With that, please join me in thanking Margaret.
We've received your request
You will be notified by email when the transcript and captions are available. The process may take up to 5 business days. Please contact firstname.lastname@example.org if you have any questions about this request.
Ever since the invention of the laser more than 50 years ago, scientists have been striving to create an X-ray version. But until recently, very high power levels were needed to make an X-ray laser. Making a practical, tabletop-scale X-ray laser source required taking a new approach, described by physicist Margaret Murnane in the Hans Bethe Lecture on Oct. 18, 2017.
“The story behind how [X-ray lasers] happened is surprising and beautiful, highlighting how powerful our ability is to manipulate nature at a quantum level,” says Murnane. These new capabilities are already impacting nano and materials science, as well as showing promise for next-generation electronics, and data and energy storage devices.
The Hans Bethe Lectures, established by the Department of Physics and the College of Arts and Sciences, honor Bethe, Cornell professor of physics from 1936 until his death in 2005. Bethe won the Nobel Prize in physics in 1967 for his description of the nuclear processes that power the sun.