When a Molecule Takes a Selfie

Paul Corkum ’67G ‘72PhD, a pioneer in ultrafast laser spectroscopy, uses lasers to generate attosecond pulses that take photographs of electrons orbiting inside atoms. In the 1980s and 1990s, he developed and confirmed models of atomic ionization and recolliding electrons. In 2001, he and his colleagues demonstrated the first laser pulses lasting less than a femtosecond. Corkum holds a joint chair in attosecond photonics at the National Research Council of Canada and the University of Ottawa. In 2009, he was elected a member of the U.S. Academy of Science and also received the Gerhard Herzberg Canada Gold Medal for Science and Engineering, the most prestigious science prize given in Canada. Corkum holds a B.S. from Acadia University in Nova Scotia and an M.S. and Ph.D. from Lehigh.

Q: Beyond scientific and mathematical intelligence, what are the qualities necessary to be a good researcher?

A: Science is just like the arts. To write a novel, for example, you need to know the skills of writing and plotting. Then you need creative ideas. Physics is no different. You need to know the basic skills of mathematics and physics. Then you need creative ideas. We see movies of how preoccupied a writer is when planning and writing a novel. It is the same in science—you must really get into the subject.

But there is one big difference. Writing is a solitary profession. Very few novels are written by a group. In contrast, physics is a very social profession. Experiments are almost always done in teams. Discussion and debate are integral parts, as are presenting and defending results to others. Rarely is a paper or a report written by an individual.

Q: What motivates you?

A: I love physics. You get to think about how nature works at its most fundamental. I love the ideas. They are so beautiful and simple! But what is most exciting is that when you have a new idea, no one in the whole of history has ever thought about it before. If they had, your idea is no longer original science—it is old news. But if you have a new idea, and it proves to be important, that is really fun. To me, coming up with a new, important idea is the intellectual equivalent to skiing down a mountain as fast as you dare.

Q: What role have Lehigh and the teachers you had here played in your career?

A: I came to Lehigh from a very small college in Nova Scotia. At Lehigh I shared an office with 4 or 5 other graduate students. They included Stephen Mack (M.S. ’67, ’71 Ph.D.), Stephen Longo (M.S. ’67), William Emkey (M.S. ’67, ’71 Ph.D.) and Thomas Tauber ’67. I remember us discussing the problems that Professors Bob Folk or Al MacLennan would assign for us to work on. These discussions went well into the night. But we discussed many other issues as well. I learned physics at Lehigh but I also grew up and learned to think critically.

Q: When did your interest in attosecond science begin?

A: I left Lehigh to move to Ottawa where I joined an experimental group at Canada’s National Research Council as a postdoc. In Ottawa we used lasers to measure the properties of plasmas. I had never seen a laser until then. Soon, I began working to make shorter light flashes to measure faster phenomena. As I got into making short pulses, I began thinking of the unique experiments that could be done with these lasers.

One day it struck me—a “blinding idea.” I knew how to make the world’s shortest pulses and I could even see how to measure them and what to do with them. It was so exciting!

So, what is the secret of maturing into a scientist with ideas? One gets so deeply involved in a subject, and so interested that you naturally think about it every time you relax. New ideas are constantly bubbling up to the surface of your mind. Of course, most of these ideas are wrong. The trick is to select the right ideas and critically reject the wrong ones as quickly as possible.

Q: How do you relate an attosecond to the speed of light? To the age of the universe?

A: This is amazing. It always impresses me when I say it. An attosecond is to one second as a second is to the age of the universe. (In fact, this slightly understates the ratio—an attosecond is a bit shorter still!). Imagine how different our world is compared to what it was 13 billion years ago. The attosecond world of electrons is equally different from our world.

You might be interested in another way to understand an attosecond. If you could travel at the speed of light—which is the fastest speed possible—you would only be able to circumnavigate (that is, to orbit) an atom in one attosecond!

Q: Why would anyone ever want such a short flash of light?

A: Everyone has seen that famous photo of a bullet exiting a light bulb that it has just pierced. We all know that the light bulb is shattered, but not yet disintegrated. To make this famous photo, H. E. Edgerton (the photographer) needed a very fast flash—a flash fast enough to freeze a speeding bullet. An attosecond pulse is much, much faster—fast enough to freeze a speeding electron. So, one can think of an attosecond pulse as a camera flash, ideally suited to measuring electrons.

Q: How did you go about modeling the ionization of atoms in the 1980s?

A: What I really did was take ideas that were well-developed in one area of physics and apply them to a new area. In this case, I applied the ideas from plasma physics that describe how intense light interacts with free electrons and ions.

I asked, “what if we were to start from an atom?” What would the resulting plasma look like? And if there were too few atoms to actually make a plasma, what characteristics would the electrons and ions have? Amazingly, nobody had thought about these questions before. The result findings led to Optical Field Ionization, a new approach to X-ray lasers, recollision and attosecond science.

Q: How significant are the OFI (optical field ionization) lasers that you helped to develop?

A: Since Einstein’s work in 1917, we have known that, if we could have an electron populate one or more high-energy levels preferentially over a lower energy level, then stimulated emission could grow into a laser. As everyone knows, in 1960 this led to the laser.

Three decades later, my colleagues and I discovered that, as a plasma recombined following OFI, high-energy electronic states filled up before lower energy ones. We immediately recognized that this would be a great way to create an X-ray laser. Scientists are still developing this idea today.

There are two alternative ways to make laser-like X-ray light that Einstein had not thought about. They are re-collision and the free-electron-laser. Neither uses bound electrons. In fact, re-collision literally breaks the electron free from the atom and exploits its early response to produce very short wavelength light – X-rays. This recollision process may sound abstract, but it is really easy to understand.

Q: Tell us about recollision.

A: Recollision is one of the simplest ideas possible. You need only think about going to the “Jersey shore.” Let me draw the analogy. Just as a water wave is a wave of force on a boat, light is a wave of force on an electron. And just as a boat moves up and down with a water wave, the electron also oscillates in a light wave.

Now imagine that the boat is moored to a dock. Similarly, the electron is confined by an atomic ion, making an atom. In the case of the boat, if a big wave comes, the rope might break and the boat is no longer confined, just as the electron can snap free from the atom using optical field ionization. Once free, the boat might smash right back into the wharf on the next swell and be shattered. Similarly, the electron can smash (“recollide”) into its parent ion. If it does crash into its parent ion, the electron gives out a flash of light—the X-ray light that we can use to make attosecond pulses. The X-ray light flash is the analogue to the crashing sound of the boat breaking up as it shatters.

Q: How did the Recollision Electron Model that you developed in the 1990s lead to the generation of attosecond pulses from lasers?

A: As soon as I understood recollision, I knew that this simple idea would lead to the most important new theory of nonlinear optics since (1981 physics Nobel laureate Nicolaas) Bloembergen. That was already very exciting. But the idea of recollision is so intuitive it was also easy to think how to apply it.

One big difference between a water wave and a light wave is that no one can control the ocean waves, so the boat is in danger for a long time. But lasers can control light waves and so we can engineer the light wave to only allow one possible crash and that crash can only last for a fraction of a light cycle. That is how we make an attosecond pulse. The current record (for shortest pulse duration), held by Professor Zenghu Chang at the University of Central Florida, is 67 attoseconds.

Q: You have said that attosecond light pulses can stimulate a molecule to “take a selfie.” Describe some of the ways this happens.

A: This is very neat and also immediately obvious. It comes from recollision. Let’s start out with a selfie. You hold the camera about ten head diameters (an arm’s length) from you and then snap the photo with the camera pointed toward you. In a CCD, the image is converted into an electrical signal, but this signal contains all information on your face. It is stored electrically and then finally transferred back to light for you to see.

Now think about a molecule that might like to take a selfie. A laser pulls an electron from the molecule to about 10 molecular diameters away and then drives it back. When it recollides, it converts to light (remember the attosecond light pulse) as the electron refills its initial place in the atom. That light contains information about the molecule’s structure. When we capture this light, it also goes to a CCD, where it is coded and stored as an electrical signal. To decode the signal, we use tomography (much like they do in the hospital). Tomography is a bit beyond the molecule’s ability, so it must call for the help of a scientist, but probably few people taking selfies understand how their camera works either.

The process I just described raises a very interesting philosophical issue: “What does the selfie really measure?” We say that it measures the difference between a molecule with all of its electrons, but with one fewer electron—the missing electron extracted by Optical Field Ionization that snaps the photo. So the electron taking the photo refills (and therefore measures) the hole it left behind. In other words, when the electron produces the light, it actually measures the wave function, not its square. The wave function (or Dyson orbital) we measured for a nitrogen molecule looks very much like the chemical orbital.

Q: Describe how you and your colleagues demonstrated experimentally the first laser pulses lasting less than a femtosecond.

A: This happened in the lab of my colleague, Ferenc Krausz (of the Max Planck Institute of Quantum Optics in Germany). He sent me the data from his experiment in the form of a draft manuscript. I had thought a lot about how to measure attosecond pulses. After I read the manuscript, I replied that his experiment was beautiful, but I disagreed with how he understood his measurement. This led to a flurry of emails across the Atlantic as we debated the issue. So in the end, I contributed the measurement method, and Krausz and his group did the experiment. It was the first-ever experiment in which an isolated pulse lasting less than one femtosecond was measured.

Q: How do you make and measure attosecond pulses?

A: The process of making them is really very mechanical and intuitive—remember the water wave analogy. The trick is to make sure that the light wave that controls the recollision electron allows only one possible crest of a wave in which the electron can be ripped from the atom and only one following crest of the wave for the electron to recollide. This is possible if the light wave has only one oscillation, or if the light wave has a special shape.

To measure attosecond pulses we begin by making a photoelectron replica of the unknown pulse. It is really the replica that we will measure. This is a good strategy because, as you will remember, light is a wave of force on an electron so once we have an electron we can use this force. For measurement we create the electron replica right in the middle of the light wave. The electron replica gets a different push from the light depending on when it is created. By measuring what happens to many electrons, we measure the duration of the pulse.

Q: What materials do these light pulses interact with?

A: An attosecond pulse is just a very short flash of ultraviolet or X-ray light. It interacts with all conventional materials – solids, liquids and gases. This is what gives attosecond science such a promising future.

Q: How will attosecond science help scientists observe and manipulate matter at the atomic scale?

A: There are several ways. Let me mention two:

• So far we have used the fact that intense light exerts a strong force on electrons. As I described, attosecond pulses are produced by using this intense light control to force a recollision. But all matter contains electrons so we can use this same property of well-controlled laser light to exercise attosecond control of bound electrons to manipulate matter. In a very exciting experiment we used intense light to spin a chlorine molecule faster and faster until the centrifugal force became strong enough to break the bond.

• Another approach uses the freeze-frame method I described when I mentioned Edgerton’s photo of a bullet: an attosecond pulse is so short that even an electron cannot move during an attosecond flash. Therefore, we can freeze-frame electron motion just as Edgerton froze the motion of a bullet, or we could take very fast movies.

Q: What new frontiers, in nanotechnology and in science in general, will attosecond science help to open?

A: We know a lot about electrons. In fact we harness them for spectroscopy for remote sensing, to make lasers and lots of other things. However, the electrons’ world has been partially obscure to us. Now it is opening. As we learn more, this could influence all of the physical sciences.

Q: Are there philosophical implications to attosecond science?

A: Let me discuss one very interesting philosophical issue: “What does the selfie really measure?” We say that it measures the difference between a molecule with all of its electrons and with one fewer electron where the missing electron is extracted by Optical Field Ionization. To create the attosecond pulse the electron taking the photo must exactly refill the hole it left when it ionized. That means we measure the wave function (up to one unknown phase) not its square. Some people have thought that wave functions are unmeasurable quantities.

Q: Have the images you’ve obtained of atoms changed our understanding of the atomic world? Of the universe and its origins?

A very famous chemist (1986 Nobel laureate Dudley Hirschbach of Harvard University) once told me that the orbital that we measured (the selfie) of a nitrogen orbital will eventually change how chemistry is taught. I am not a chemist, so I cannot judge if this is correct. As I understand it, elementary chemistry introduces orbitals abstractly. You can understand why—there is really no such thing as a single electron occupying a molecular orbital. A molecule’s electrons are constantly interacting with each other and interchanging roles. Really an orbital is a mathematical construct that is very useful for understanding chemical reactions. Now that we can “take a selfie,” we can say that there is an experimental way to see an orbital. We do it by extracting an electron first so we can isolate it from the others and then we experimentally observe the hole it left behind.

You ask about the universe. Although an attosecond is a very short time, we are nowhere near a physical limit. The world of the atomic nucleus will have processes that are much faster than we can measure at present, and the sub-nuclear world is faster still. So there are lots of opportunities for bright students but for the present, the science of fast measurement is still confined to normal matter.